View Source Evision (Evision v0.1.37)

Summary

Functions

Calculates the per-element absolute difference between two arrays or between an array and a scalar.

Calculates the per-element absolute difference between two arrays or between an array and a scalar.

Adds an image to the accumulator image.

Adds an image to the accumulator image.

Adds the per-element product of two input images to the accumulator image.

Adds the per-element product of two input images to the accumulator image.

Adds the square of a source image to the accumulator image.

Adds the square of a source image to the accumulator image.

Updates a running average.

Updates a running average.

Calculates the per-element sum of two arrays or an array and a scalar.

Calculates the per-element sum of two arrays or an array and a scalar.

Draws a text on the image.

Draws a text on the image.

Calculates the weighted sum of two arrays.

Calculates the weighted sum of two arrays.

Variant 1:

Applies a user colormap on a given image.

Variant 1:

Applies a user colormap on a given image.

Approximates a polygonal curve(s) with the specified precision.

Approximates a polygonal curve(s) with the specified precision.

Calculates a contour perimeter or a curve length.

Draws an arrow segment pointing from the first point to the second one.

Draws an arrow segment pointing from the first point to the second one.

naive nearest neighbor finder

naive nearest neighbor finder

Applies the bilateral filter to an image.

Applies the bilateral filter to an image.

Blurs an image using the normalized box filter.

Blurs an image using the normalized box filter.

Computes the source location of an extrapolated pixel.

Calculates the up-right bounding rectangle of a point set or non-zero pixels of gray-scale image.

Blurs an image using the box filter.

Blurs an image using the box filter.

Finds the four vertices of a rotated rect. Useful to draw the rotated rectangle.

Finds the four vertices of a rotated rect. Useful to draw the rotated rectangle.

Broadcast the given Mat to the given shape.

Broadcast the given Mat to the given shape.

Constructs the image pyramid which can be passed to calcOpticalFlowPyrLK.

Constructs the image pyramid which can be passed to calcOpticalFlowPyrLK.

Calculates an optical flow for a sparse feature set using the iterative Lucas-Kanade method with pyramids.

Calculates an optical flow for a sparse feature set using the iterative Lucas-Kanade method with pyramids.

Finds the camera intrinsic and extrinsic parameters from several views of a calibration pattern.

Finds the camera intrinsic and extrinsic parameters from several views of a calibration pattern.

Finds the camera intrinsic and extrinsic parameters from several views of a calibration pattern.

Finds the camera intrinsic and extrinsic parameters from several views of a calibration pattern.

Computes Hand-Eye calibration: \f$_{}^{g}\textrm{T}_c\f$

Computes Hand-Eye calibration: \f$_{}^{g}\textrm{T}_c\f$

Computes Robot-World/Hand-Eye calibration: \f$_{}^{w}\textrm{T}_b\f$ and \f$_{}^{c}\textrm{T}_g\f$

Computes Robot-World/Hand-Eye calibration: \f$_{}^{w}\textrm{T}_b\f$ and \f$_{}^{c}\textrm{T}_g\f$

Computes useful camera characteristics from the camera intrinsic matrix.

Finds an object center, size, and orientation.

Finds edges in an image using the Canny algorithm @cite Canny86 .

Calculates the magnitude and angle of 2D vectors.

Calculates the magnitude and angle of 2D vectors.

checkChessboard

Returns true if the specified feature is supported by the host hardware.

Checks every element of an input array for invalid values.

Checks every element of an input array for invalid values.

Given an original color image, two differently colored versions of this image can be mixed seamlessly.

Given an original color image, two differently colored versions of this image can be mixed seamlessly.

Performs the per-element comparison of two arrays or an array and scalar value.

Performs the per-element comparison of two arrays or an array and scalar value.

Compares two histograms.

Copies the lower or the upper half of a square matrix to its another half.

Copies the lower or the upper half of a square matrix to its another half.

Combines two rotation-and-shift transformations.

Combines two rotation-and-shift transformations.

For points in an image of a stereo pair, computes the corresponding epilines in the other image.

For points in an image of a stereo pair, computes the corresponding epilines in the other image.

Computes the Enhanced Correlation Coefficient value between two images @cite EP08 .

Computes the Enhanced Correlation Coefficient value between two images @cite EP08 .

connectedComponents

connectedComponents

computes the connected components labeled image of boolean image

computes the connected components labeled image of boolean image

connectedComponentsWithStats

connectedComponentsWithStats

computes the connected components labeled image of boolean image and also produces a statistics output for each label

computes the connected components labeled image of boolean image and also produces a statistics output for each label

Calculates a contour area.

Calculates a contour area.

Converts an array to half precision floating number.

Converts an array to half precision floating number.

Converts image transformation maps from one representation to another.

Converts image transformation maps from one representation to another.

Converts points from homogeneous to Euclidean space.

Converts points from homogeneous to Euclidean space.

Converts points from Euclidean to homogeneous space.

Converts points from Euclidean to homogeneous space.

Scales, calculates absolute values, and converts the result to 8-bit.

Scales, calculates absolute values, and converts the result to 8-bit.

Finds the convex hull of a point set.

Finds the convex hull of a point set.

Finds the convexity defects of a contour.

Finds the convexity defects of a contour.

This is an overloaded member function, provided for convenience (python) Copies the matrix to another one. When the operation mask is specified, if the Mat::create call shown above reallocates the matrix, the newly allocated matrix is initialized with all zeros before copying the data.

This is an overloaded member function, provided for convenience (python) Copies the matrix to another one. When the operation mask is specified, if the Mat::create call shown above reallocates the matrix, the newly allocated matrix is initialized with all zeros before copying the data.

Calculates eigenvalues and eigenvectors of image blocks for corner detection.

Calculates eigenvalues and eigenvectors of image blocks for corner detection.

Harris corner detector.

Calculates the minimal eigenvalue of gradient matrices for corner detection.

Calculates the minimal eigenvalue of gradient matrices for corner detection.

Refines coordinates of corresponding points.

Refines coordinates of corresponding points.

Counts non-zero array elements.

createAffineTransformer

Creates AlignMTB object

Creates AlignMTB object

Creates KNN Background Subtractor

Creates KNN Background Subtractor

Creates MOG2 Background Subtractor

Creates MOG2 Background Subtractor

Creates CalibrateDebevec object

Creates CalibrateDebevec object

Creates CalibrateRobertson object

Creates CalibrateRobertson object

createChiHistogramCostExtractor

createChiHistogramCostExtractor

Creates a smart pointer to a cv::CLAHE class and initializes it.

Creates a smart pointer to a cv::CLAHE class and initializes it.

createEMDHistogramCostExtractor

createEMDHistogramCostExtractor

createEMDL1HistogramCostExtractor

createEMDL1HistogramCostExtractor

Creates a smart pointer to a cv::GeneralizedHoughBallard class and initializes it.

Creates a smart pointer to a cv::GeneralizedHoughGuil class and initializes it.

This function computes a Hanning window coefficients in two dimensions.

This function computes a Hanning window coefficients in two dimensions.

createHausdorffDistanceExtractor

createHausdorffDistanceExtractor

Creates a smart pointer to a LineSegmentDetector object and initializes it.

Creates a smart pointer to a LineSegmentDetector object and initializes it.

Creates MergeDebevec object

Creates MergeMertens object

Creates MergeMertens object

Creates MergeRobertson object

createNormHistogramCostExtractor

createNormHistogramCostExtractor

createShapeContextDistanceExtractor

createShapeContextDistanceExtractor

createThinPlateSplineShapeTransformer

createThinPlateSplineShapeTransformer

Creates simple linear mapper with gamma correction

Creates simple linear mapper with gamma correction

Creates TonemapDrago object

Creates TonemapDrago object

Creates TonemapMantiuk object

Creates TonemapMantiuk object

Creates TonemapReinhard object

Creates TonemapReinhard object

Computes the cube root of an argument.

Converts an image from one color space to another.

Converts an image from one color space to another.

Converts an image from one color space to another where the source image is stored in two planes.

Converts an image from one color space to another where the source image is stored in two planes.

Performs a forward or inverse discrete Cosine transform of 1D or 2D array.

Performs a forward or inverse discrete Cosine transform of 1D or 2D array.

Transforms a color image to a grayscale image. It is a basic tool in digital printing, stylized black-and-white photograph rendering, and in many single channel image processing applications

Transforms a color image to a grayscale image. It is a basic tool in digital printing, stylized black-and-white photograph rendering, and in many single channel image processing applications

Decompose an essential matrix to possible rotations and translation.

Decompose an essential matrix to possible rotations and translation.

Decompose a homography matrix to rotation(s), translation(s) and plane normal(s).

Decompose a homography matrix to rotation(s), translation(s) and plane normal(s).

Decomposes a projection matrix into a rotation matrix and a camera intrinsic matrix.

Decomposes a projection matrix into a rotation matrix and a camera intrinsic matrix.

main function for all demosaicing processes

main function for all demosaicing processes

This filter enhances the details of a particular image.

This filter enhances the details of a particular image.

Returns the determinant of a square floating-point matrix.

Performs a forward or inverse Discrete Fourier transform of a 1D or 2D floating-point array.

Performs a forward or inverse Discrete Fourier transform of a 1D or 2D floating-point array.

Dilates an image by using a specific structuring element.

Dilates an image by using a specific structuring element.

Displays a text on a window image as an overlay for a specified duration.

Displays a text on a window image as an overlay for a specified duration.

Displays a text on the window statusbar during the specified period of time.

Displays a text on the window statusbar during the specified period of time.

Calculates the distance to the closest zero pixel for each pixel of the source image.

Calculates the distance to the closest zero pixel for each pixel of the source image.

Variant 1:

divide

Variant 1:

divide

Performs the per-element division of the first Fourier spectrum by the second Fourier spectrum.

Performs the per-element division of the first Fourier spectrum by the second Fourier spectrum.

Draws contours outlines or filled contours.

Draws contours outlines or filled contours.

Draw axes of the world/object coordinate system from pose estimation. @sa solvePnP

Draw axes of the world/object coordinate system from pose estimation. @sa solvePnP

Draws a marker on a predefined position in an image.

Draws a marker on a predefined position in an image.

Draws the found matches of keypoints from two images.

Filtering is the fundamental operation in image and video processing. Edge-preserving smoothing filters are used in many different applications @cite EM11 .

Filtering is the fundamental operation in image and video processing. Edge-preserving smoothing filters are used in many different applications @cite EM11 .

Calculates eigenvalues and eigenvectors of a symmetric matrix.

Calculates eigenvalues and eigenvectors of a symmetric matrix.

Calculates eigenvalues and eigenvectors of a non-symmetric matrix (real eigenvalues only).

Calculates eigenvalues and eigenvectors of a non-symmetric matrix (real eigenvalues only).

Approximates an elliptic arc with a polyline.

Draws a simple or thick elliptic arc or fills an ellipse sector.

Draws a simple or thick elliptic arc or fills an ellipse sector.

Computes the "minimal work" distance between two weighted point configurations.

Computes the "minimal work" distance between two weighted point configurations.

return a list of enabled modules in this build

Equalizes the histogram of a grayscale image.

Equalizes the histogram of a grayscale image.

Erodes an image by using a specific structuring element.

Erodes an image by using a specific structuring element.

Computes an optimal affine transformation between two 2D point sets.

Variant 1:

estimateAffine2D

Computes an optimal affine transformation between two 3D point sets.

Computes an optimal affine transformation between two 3D point sets.

Computes an optimal limited affine transformation with 4 degrees of freedom between two 2D point sets.

Computes an optimal limited affine transformation with 4 degrees of freedom between two 2D point sets.

Estimates the sharpness of a detected chessboard.

Estimates the sharpness of a detected chessboard.

Computes an optimal translation between two 3D point sets.

Computes an optimal translation between two 3D point sets.

Calculates the exponent of every array element.

Calculates the exponent of every array element.

Extracts a single channel from src (coi is 0-based index)

Extracts a single channel from src (coi is 0-based index)

Calculates the angle of a 2D vector in degrees.

Perform image denoising using Non-local Means Denoising algorithm http://www.ipol.im/pub/algo/bcm_non_local_means_denoising/ with several computational optimizations. Noise expected to be a gaussian white noise

Variant 1:

Perform image denoising using Non-local Means Denoising algorithm http://www.ipol.im/pub/algo/bcm_non_local_means_denoising/ with several computational optimizations. Noise expected to be a gaussian white noise

Perform image denoising using Non-local Means Denoising algorithm http://www.ipol.im/pub/algo/bcm_non_local_means_denoising/ with several computational optimizations. Noise expected to be a gaussian white noise

Modification of fastNlMeansDenoising function for colored images

Modification of fastNlMeansDenoising function for colored images

Modification of fastNlMeansDenoisingMulti function for colored images sequences

Modification of fastNlMeansDenoisingMulti function for colored images sequences

Modification of fastNlMeansDenoising function for images sequence where consecutive images have been captured in small period of time. For example video. This version of the function is for grayscale images or for manual manipulation with colorspaces. See @cite Buades2005DenoisingIS for more details (open access here).

Variant 1:

Modification of fastNlMeansDenoising function for images sequence where consecutive images have been captured in small period of time. For example video. This version of the function is for grayscale images or for manual manipulation with colorspaces. See @cite Buades2005DenoisingIS for more details (open access here).

Modification of fastNlMeansDenoising function for images sequence where consecutive images have been captured in small period of time. For example video. This version of the function is for grayscale images or for manual manipulation with colorspaces. See @cite Buades2005DenoisingIS for more details (open access here).

Fills a convex polygon.

Fills the area bounded by one or more polygons.

Fills the area bounded by one or more polygons.

Convolves an image with the kernel.

Convolves an image with the kernel.

Filters homography decompositions based on additional information.

Filters homography decompositions based on additional information.

Filters off small noise blobs (speckles) in the disparity map

Filters off small noise blobs (speckles) in the disparity map

Finds the positions of internal corners of the chessboard.

Finds the positions of internal corners of the chessboard.

Finds the positions of internal corners of the chessboard using a sector based approach.

Finds the positions of internal corners of the chessboard using a sector based approach.

Finds contours in a binary image.

Finds contours in a binary image.

Variant 1:

Calculates an essential matrix from the corresponding points in two images.

Calculates an essential matrix from the corresponding points in two images.

Calculates an essential matrix from the corresponding points in two images from potentially two different cameras.

Variant 1:

findFundamentalMat

Calculates a fundamental matrix from the corresponding points in two images.

Calculates a fundamental matrix from the corresponding points in two images.

Finds a perspective transformation between two planes.

Returns the list of locations of non-zero pixels

Returns the list of locations of non-zero pixels

Finds the geometric transform (warp) between two images in terms of the ECC criterion @cite EP08 .

Fits an ellipse around a set of 2D points.

Fits an ellipse around a set of 2D points.

Fits an ellipse around a set of 2D points.

Fits a line to a 2D or 3D point set.

Fits a line to a 2D or 3D point set.

Flips a 2D array around vertical, horizontal, or both axes.

Flips a 2D array around vertical, horizontal, or both axes.

Flips a n-dimensional at given axis

Flips a n-dimensional at given axis

Fills a connected component with the given color.

Fills a connected component with the given color.

Blurs an image using a Gaussian filter.

Blurs an image using a Gaussian filter.

Performs generalized matrix multiplication.

Performs generalized matrix multiplication.

getAffineTransform

Returns full configuration time cmake output.

Returns list of CPU features enabled during compilation.

Returns the number of CPU ticks.

Returns the default new camera matrix.

Returns the default new camera matrix.

Returns filter coefficients for computing spatial image derivatives.

Returns filter coefficients for computing spatial image derivatives.

Calculates the font-specific size to use to achieve a given height in pixels.

Calculates the font-specific size to use to achieve a given height in pixels.

Returns Gabor filter coefficients.

Returns Gabor filter coefficients.

Returns Gaussian filter coefficients.

Returns Gaussian filter coefficients.

Returns feature name by ID

getLogLevel

Returns the number of logical CPUs available for the process.

Returns the number of threads used by OpenCV for parallel regions.

Returns the optimal DFT size for a given vector size.

Returns the new camera intrinsic matrix based on the free scaling parameter.

Returns the new camera intrinsic matrix based on the free scaling parameter.

Calculates a perspective transform from four pairs of the corresponding points.

Calculates a perspective transform from four pairs of the corresponding points.

Retrieves a pixel rectangle from an image with sub-pixel accuracy.

Retrieves a pixel rectangle from an image with sub-pixel accuracy.

Calculates an affine matrix of 2D rotation.

Returns a structuring element of the specified size and shape for morphological operations.

Returns a structuring element of the specified size and shape for morphological operations.

Calculates the width and height of a text string.

Returns the index of the currently executed thread within the current parallel region. Always returns 0 if called outside of parallel region.

Returns the number of ticks.

Returns the number of ticks per second.

Returns the trackbar position.

Returns major library version

Returns minor library version

Returns revision field of the library version

Returns library version string

Provides rectangle of image in the window.

Provides parameters of a window.

Same as above, but returns also quality measure of the detected corners.

Same as above, but returns also quality measure of the detected corners.

Checks for the presence of at least one non-zero array element.

Returns true if the specified image can be decoded by OpenCV

Returns true if an image with the specified filename can be encoded by OpenCV

haveOpenVX

hconcat

Finds circles in a grayscale image using the Hough transform.

Finds circles in a grayscale image using the Hough transform.

Finds lines in a binary image using the standard Hough transform.

Finds lines in a binary image using the standard Hough transform.

Finds line segments in a binary image using the probabilistic Hough transform.

Finds line segments in a binary image using the probabilistic Hough transform.

Finds lines in a binary image using the standard Hough transform and get accumulator.

Finds lines in a binary image using the standard Hough transform and get accumulator.

HuMoments

HuMoments

Calculates the inverse Discrete Cosine Transform of a 1D or 2D array.

Calculates the inverse Discrete Cosine Transform of a 1D or 2D array.

Calculates the inverse Discrete Fourier Transform of a 1D or 2D array.

Calculates the inverse Discrete Fourier Transform of a 1D or 2D array.

Applying an appropriate non-linear transformation to the gradient field inside the selection and then integrating back with a Poisson solver, modifies locally the apparent illumination of an image.

Applying an appropriate non-linear transformation to the gradient field inside the selection and then integrating back with a Poisson solver, modifies locally the apparent illumination of an image.

Returns the number of images inside the give file

Returns the number of images inside the give file

Reads a multi-page image from a buffer in memory.

Reads a multi-page image from a buffer in memory.

Encodes an image into a memory buffer.

Encodes an image into a memory buffer.

Loads an image from a file.

Loads an image from a file.

Loads a multi-page image from a file.

Loads a multi-page image from a file.

Loads a of images of a multi-page image from a file.

Loads a of images of a multi-page image from a file.

Saves an image to a specified file.

Saves an image to a specified file.

Finds an initial camera intrinsic matrix from 3D-2D point correspondences.

Finds an initial camera intrinsic matrix from 3D-2D point correspondences.

Computes the projection and inverse-rectification transformation map. In essense, this is the inverse of #initUndistortRectifyMap to accomodate stereo-rectification of projectors ('inverse-cameras') in projector-camera pairs.

Computes the projection and inverse-rectification transformation map. In essense, this is the inverse of #initUndistortRectifyMap to accomodate stereo-rectification of projectors ('inverse-cameras') in projector-camera pairs.

Computes the undistortion and rectification transformation map.

Computes the undistortion and rectification transformation map.

Restores the selected region in an image using the region neighborhood.

Restores the selected region in an image using the region neighborhood.

Checks if array elements lie between the elements of two other arrays.

Checks if array elements lie between the elements of two other arrays.

Inserts a single channel to dst (coi is 0-based index)

integral2

Calculates the integral of an image.

Calculates the integral of an image.

integral

Finds intersection of two convex polygons

Finds intersection of two convex polygons

Finds the inverse or pseudo-inverse of a matrix.

Finds the inverse or pseudo-inverse of a matrix.

Inverts an affine transformation.

Inverts an affine transformation.

Tests a contour convexity.

Finds centers of clusters and groups input samples around the clusters.

Finds centers of clusters and groups input samples around the clusters.

Calculates the Laplacian of an image.

Calculates the Laplacian of an image.

Draws a line segment connecting two points.

Draws a line segment connecting two points.

Remaps an image to polar coordinates space.

Remaps an image to polar coordinates space.

Calculates the natural logarithm of every array element.

Calculates the natural logarithm of every array element.

Remaps an image to semilog-polar coordinates space.

Remaps an image to semilog-polar coordinates space.

Performs a look-up table transform of an array.

Performs a look-up table transform of an array.

Calculates the magnitude of 2D vectors.

Calculates the magnitude of 2D vectors.

Calculates the Mahalanobis distance between two vectors.

Compares a template against overlapped image regions.

Compares a template against overlapped image regions.

Computes partial derivatives of the matrix product for each multiplied matrix.

Computes partial derivatives of the matrix product for each multiplied matrix.

Calculates per-element maximum of two arrays or an array and a scalar.

Calculates per-element maximum of two arrays or an array and a scalar.

Calculates an average (mean) of array elements.

Calculates an average (mean) of array elements.

Finds an object on a back projection image.

meanStdDev

Blurs an image using the median filter.

Blurs an image using the median filter.

merge

Calculates per-element minimum of two arrays or an array and a scalar.

Calculates per-element minimum of two arrays or an array and a scalar.

Finds a rotated rectangle of the minimum area enclosing the input 2D point set.

Finds a circle of the minimum area enclosing a 2D point set.

Finds a triangle of minimum area enclosing a 2D point set and returns its area.

Finds a triangle of minimum area enclosing a 2D point set and returns its area.

Finds the global minimum and maximum in an array.

Finds the global minimum and maximum in an array.

Calculates all of the moments up to the third order of a polygon or rasterized shape.

Calculates all of the moments up to the third order of a polygon or rasterized shape.

Performs advanced morphological transformations.

Performs advanced morphological transformations.

Moves the window to the specified position

Performs the per-element multiplication of two Fourier spectrums.

Performs the per-element multiplication of two Fourier spectrums.

Calculates the product of a matrix and its transposition.

Calculates the product of a matrix and its transposition.

Calculates the per-element scaled product of two arrays.

Calculates the per-element scaled product of two arrays.

Creates a window.

Creates a window.

Calculates the absolute norm of an array.

Variant 1:

Calculates an absolute difference norm or a relative difference norm.

Calculates an absolute difference norm or a relative difference norm.

Normalizes the norm or value range of an array.

Normalizes the norm or value range of an array.

Replaces NaNs by given number

Replaces NaNs by given number

Variant 1:

PCACompute2

Variant 1:

PCACompute

Pencil-like non-photorealistic line drawing

Pencil-like non-photorealistic line drawing

Performs the perspective matrix transformation of vectors.

Performs the perspective matrix transformation of vectors.

Calculates the rotation angle of 2D vectors.

Calculates the rotation angle of 2D vectors.

The function is used to detect translational shifts that occur between two images.

The function is used to detect translational shifts that occur between two images.

Performs a point-in-contour test.

Calculates x and y coordinates of 2D vectors from their magnitude and angle.

Calculates x and y coordinates of 2D vectors from their magnitude and angle.

Polls for a pressed key.

Draws several polygonal curves.

Draws several polygonal curves.

Raises every array element to a power.

Raises every array element to a power.

Calculates a feature map for corner detection.

Calculates a feature map for corner detection.

Computes the Peak Signal-to-Noise Ratio (PSNR) image quality metric.

Computes the Peak Signal-to-Noise Ratio (PSNR) image quality metric.

Blurs an image and downsamples it.

Blurs an image and downsamples it.

Performs initial step of meanshift segmentation of an image.

Performs initial step of meanshift segmentation of an image.

Upsamples an image and then blurs it.

Upsamples an image and then blurs it.

Fills the array with normally distributed random numbers.

Shuffles the array elements randomly.

Shuffles the array elements randomly.

Generates a single uniformly-distributed random number or an array of random numbers.

Read a .flo file

Variant 1:

Recovers the relative camera rotation and the translation from an estimated essential matrix and the corresponding points in two images, using chirality check. Returns the number of inliers that pass the check.

Variant 1:

Recovers the relative camera rotation and the translation from corresponding points in two images from two different cameras, using cheirality check. Returns the number of inliers that pass the check.

Recovers the relative camera rotation and the translation from corresponding points in two images from two different cameras, using cheirality check. Returns the number of inliers that pass the check.

Variant 1:

Draws a simple, thick, or filled up-right rectangle.

Draws a simple, thick, or filled up-right rectangle.

Finds out if there is any intersection between two rectangles

Reduces a matrix to a vector.

Reduces a matrix to a vector.

Finds indices of max elements along provided axis

Finds indices of max elements along provided axis

Finds indices of min elements along provided axis

Finds indices of min elements along provided axis

Applies a generic geometrical transformation to an image.

Applies a generic geometrical transformation to an image.

Fills the output array with repeated copies of the input array.

Fills the output array with repeated copies of the input array.

Reprojects a disparity image to 3D space.

Reprojects a disparity image to 3D space.

Resizes an image.

Resizes an image.

Resizes the window to the specified size

Converts a rotation matrix to a rotation vector or vice versa.

Converts a rotation matrix to a rotation vector or vice versa.

Rotates a 2D array in multiples of 90 degrees. The function cv::rotate rotates the array in one of three different ways: Rotate by 90 degrees clockwise (rotateCode = ROTATE_90_CLOCKWISE). Rotate by 180 degrees clockwise (rotateCode = ROTATE_180). Rotate by 270 degrees clockwise (rotateCode = ROTATE_90_COUNTERCLOCKWISE).

Rotates a 2D array in multiples of 90 degrees. The function cv::rotate rotates the array in one of three different ways: Rotate by 90 degrees clockwise (rotateCode = ROTATE_90_CLOCKWISE). Rotate by 180 degrees clockwise (rotateCode = ROTATE_180). Rotate by 270 degrees clockwise (rotateCode = ROTATE_90_COUNTERCLOCKWISE).

Finds out if there is any intersection between two rotated rectangles.

Finds out if there is any intersection between two rotated rectangles.

Computes an RQ decomposition of 3x3 matrices.

Computes an RQ decomposition of 3x3 matrices.

Calculates the Sampson Distance between two points.

Calculates the sum of a scaled array and another array.

Calculates the sum of a scaled array and another array.

Calculates the first x- or y- image derivative using Scharr operator.

Calculates the first x- or y- image derivative using Scharr operator.

Image editing tasks concern either global changes (color/intensity corrections, filters, deformations) or local changes concerned to a selection. Here we are interested in achieving local changes, ones that are restricted to a region manually selected (ROI), in a seamless and effortless manner. The extent of the changes ranges from slight distortions to complete replacement by novel content @cite PM03 .

Image editing tasks concern either global changes (color/intensity corrections, filters, deformations) or local changes concerned to a selection. Here we are interested in achieving local changes, ones that are restricted to a region manually selected (ROI), in a seamless and effortless manner. The extent of the changes ranges from slight distortions to complete replacement by novel content @cite PM03 .

selectROI

Variant 1:

Allows users to select a ROI on the given image.

Allows users to select a ROI on the given image.

Allows users to select multiple ROIs on the given image.

Allows users to select multiple ROIs on the given image.

Applies a separable linear filter to an image.

Applies a separable linear filter to an image.

Initializes a scaled identity matrix.

Initializes a scaled identity matrix.

setLogLevel

OpenCV will try to set the number of threads for subsequent parallel regions.

Sets state of default random number generator.

Sets the trackbar maximum position.

Sets the trackbar minimum position.

Sets the trackbar position.

setUseOpenVX

Enables or disables the optimized code.

Changes parameters of a window dynamically.

Updates window title

Calculates the first, second, third, or mixed image derivatives using an extended Sobel operator.

Calculates the first, second, third, or mixed image derivatives using an extended Sobel operator.

Solves one or more linear systems or least-squares problems.

Solves one or more linear systems or least-squares problems.

Finds the real roots of a cubic equation.

Finds the real roots of a cubic equation.

Variant 1:

Solve given (non-integer) linear programming problem using the Simplex Algorithm (Simplex Method).

Solve given (non-integer) linear programming problem using the Simplex Algorithm (Simplex Method).

Finds an object pose from 3 3D-2D point correspondences.

Finds an object pose from 3 3D-2D point correspondences.

Finds an object pose from 3D-2D point correspondences.

Finds an object pose from 3D-2D point correspondences.

Finds an object pose from 3D-2D point correspondences.

Finds an object pose from 3D-2D point correspondences.

Refine a pose (the translation and the rotation that transform a 3D point expressed in the object coordinate frame to the camera coordinate frame) from a 3D-2D point correspondences and starting from an initial solution.

Refine a pose (the translation and the rotation that transform a 3D point expressed in the object coordinate frame to the camera coordinate frame) from a 3D-2D point correspondences and starting from an initial solution.

Refine a pose (the translation and the rotation that transform a 3D point expressed in the object coordinate frame to the camera coordinate frame) from a 3D-2D point correspondences and starting from an initial solution.

Refine a pose (the translation and the rotation that transform a 3D point expressed in the object coordinate frame to the camera coordinate frame) from a 3D-2D point correspondences and starting from an initial solution.

Finds the real or complex roots of a polynomial equation.

Finds the real or complex roots of a polynomial equation.

Sorts each row or each column of a matrix.

Sorts each row or each column of a matrix.

Sorts each row or each column of a matrix.

Sorts each row or each column of a matrix.

Calculates the first order image derivative in both x and y using a Sobel operator

Calculates the first order image derivative in both x and y using a Sobel operator

split

Calculates the normalized sum of squares of the pixel values overlapping the filter.

Calculates the normalized sum of squares of the pixel values overlapping the filter.

Calculates a square root of array elements.

Calculates a square root of array elements.

Blurs an image using the stackBlur.

Blurs an image using the stackBlur.

startWindowThread

Calibrates a stereo camera set up. This function finds the intrinsic parameters for each of the two cameras and the extrinsic parameters between the two cameras.

Calibrates a stereo camera set up. This function finds the intrinsic parameters for each of the two cameras and the extrinsic parameters between the two cameras.

Computes rectification transforms for each head of a calibrated stereo camera.

Computes rectification transforms for each head of a calibrated stereo camera.

Computes a rectification transform for an uncalibrated stereo camera.

Computes a rectification transform for an uncalibrated stereo camera.

Stylization aims to produce digital imagery with a wide variety of effects not focused on photorealism. Edge-aware filters are ideal for stylization, as they can abstract regions of low contrast while preserving, or enhancing, high-contrast features.

Stylization aims to produce digital imagery with a wide variety of effects not focused on photorealism. Edge-aware filters are ideal for stylization, as they can abstract regions of low contrast while preserving, or enhancing, high-contrast features.

Calculates the per-element difference between two arrays or array and a scalar.

Calculates the per-element difference between two arrays or array and a scalar.

Calculates the sum of array elements.

SVDecomp

By retaining only the gradients at edge locations, before integrating with the Poisson solver, one washes out the texture of the selected region, giving its contents a flat aspect. Here Canny Edge %Detector is used.

By retaining only the gradients at edge locations, before integrating with the Poisson solver, one washes out the texture of the selected region, giving its contents a flat aspect. Here Canny Edge %Detector is used.

Applies a fixed-level threshold to each array element.

Applies a fixed-level threshold to each array element.

Returns the trace of a matrix.

Performs the matrix transformation of every array element.

Performs the matrix transformation of every array element.

Transposes a matrix.

Transposes a matrix.

Transpose for n-dimensional matrices.

Transpose for n-dimensional matrices.

This function reconstructs 3-dimensional points (in homogeneous coordinates) by using their observations with a stereo camera.

This function reconstructs 3-dimensional points (in homogeneous coordinates) by using their observations with a stereo camera.

Transforms an image to compensate for lens distortion.

Transforms an image to compensate for lens distortion.

Compute undistorted image points position

Compute undistorted image points position

Computes the ideal point coordinates from the observed point coordinates.

Computes the ideal point coordinates from the observed point coordinates.

useOpenVX

Returns the status of optimized code usage.

vconcat

Similar to #waitKey, but returns full key code.

Similar to #waitKey, but returns full key code.

Applies an affine transformation to an image.

Applies an affine transformation to an image.

Applies a perspective transformation to an image.

Applies a perspective transformation to an image.

Performs a marker-based image segmentation using the watershed algorithm.

Write a .flo to disk

Functions

Calculates the per-element absolute difference between two arrays or between an array and a scalar.

Positional Arguments
  • src1: Evision.Mat.t().

    first input array or a scalar.

  • src2: Evision.Mat.t().

    second input array or a scalar.

Return
  • dst: Evision.Mat.t().

    output array that has the same size and type as input arrays.

The function cv::absdiff calculates: Absolute difference between two arrays when they have the same size and type: \f[\texttt{dst}(I) = \texttt{saturate} (| \texttt{src1}(I) - \texttt{src2}(I)|)\f] Absolute difference between an array and a scalar when the second array is constructed from Scalar or has as many elements as the number of channels in src1: \f[\texttt{dst}(I) = \texttt{saturate} (| \texttt{src1}(I) - \texttt{src2} |)\f] Absolute difference between a scalar and an array when the first array is constructed from Scalar or has as many elements as the number of channels in src2: \f[\texttt{dst}(I) = \texttt{saturate} (| \texttt{src1} - \texttt{src2}(I) |)\f] where I is a multi-dimensional index of array elements. In case of multi-channel arrays, each channel is processed independently. Note: Saturation is not applied when the arrays have the depth CV_32S. You may even get a negative value in the case of overflow. Note: (Python) Be careful to difference behaviour between src1/src2 are single number and they are tuple/array. absdiff(src,X) means absdiff(src,(X,X,X,X)). absdiff(src,(X,)) means absdiff(src,(X,0,0,0)). @sa cv::abs(const Mat&)

Python prototype (for reference only):

absdiff(src1, src2[, dst]) -> dst
Link to this function

absdiff(src1, src2, opts)

View Source
@spec absdiff(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Calculates the per-element absolute difference between two arrays or between an array and a scalar.

Positional Arguments
  • src1: Evision.Mat.t().

    first input array or a scalar.

  • src2: Evision.Mat.t().

    second input array or a scalar.

Return
  • dst: Evision.Mat.t().

    output array that has the same size and type as input arrays.

The function cv::absdiff calculates: Absolute difference between two arrays when they have the same size and type: \f[\texttt{dst}(I) = \texttt{saturate} (| \texttt{src1}(I) - \texttt{src2}(I)|)\f] Absolute difference between an array and a scalar when the second array is constructed from Scalar or has as many elements as the number of channels in src1: \f[\texttt{dst}(I) = \texttt{saturate} (| \texttt{src1}(I) - \texttt{src2} |)\f] Absolute difference between a scalar and an array when the first array is constructed from Scalar or has as many elements as the number of channels in src2: \f[\texttt{dst}(I) = \texttt{saturate} (| \texttt{src1} - \texttt{src2}(I) |)\f] where I is a multi-dimensional index of array elements. In case of multi-channel arrays, each channel is processed independently. Note: Saturation is not applied when the arrays have the depth CV_32S. You may even get a negative value in the case of overflow. Note: (Python) Be careful to difference behaviour between src1/src2 are single number and they are tuple/array. absdiff(src,X) means absdiff(src,(X,X,X,X)). absdiff(src,(X,)) means absdiff(src,(X,0,0,0)). @sa cv::abs(const Mat&)

Python prototype (for reference only):

absdiff(src1, src2[, dst]) -> dst
@spec accumulate(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

Adds an image to the accumulator image.

Positional Arguments
  • src: Evision.Mat.t().

    Input image of type CV_8UC(n), CV_16UC(n), CV_32FC(n) or CV_64FC(n), where n is a positive integer.

Keyword Arguments
  • mask: Evision.Mat.t().

    Optional operation mask.

Return
  • dst: Evision.Mat.t().

    %Accumulator image with the same number of channels as input image, and a depth of CV_32F or CV_64F.

The function adds src or some of its elements to dst : \f[\texttt{dst} (x,y) \leftarrow \texttt{dst} (x,y) + \texttt{src} (x,y) \quad \text{if} \quad \texttt{mask} (x,y) \ne 0\f] The function supports multi-channel images. Each channel is processed independently. The function cv::accumulate can be used, for example, to collect statistics of a scene background viewed by a still camera and for the further foreground-background segmentation.

@sa accumulateSquare, accumulateProduct, accumulateWeighted

Python prototype (for reference only):

accumulate(src, dst[, mask]) -> dst
Link to this function

accumulate(src, dst, opts)

View Source
@spec accumulate(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Adds an image to the accumulator image.

Positional Arguments
  • src: Evision.Mat.t().

    Input image of type CV_8UC(n), CV_16UC(n), CV_32FC(n) or CV_64FC(n), where n is a positive integer.

Keyword Arguments
  • mask: Evision.Mat.t().

    Optional operation mask.

Return
  • dst: Evision.Mat.t().

    %Accumulator image with the same number of channels as input image, and a depth of CV_32F or CV_64F.

The function adds src or some of its elements to dst : \f[\texttt{dst} (x,y) \leftarrow \texttt{dst} (x,y) + \texttt{src} (x,y) \quad \text{if} \quad \texttt{mask} (x,y) \ne 0\f] The function supports multi-channel images. Each channel is processed independently. The function cv::accumulate can be used, for example, to collect statistics of a scene background viewed by a still camera and for the further foreground-background segmentation.

@sa accumulateSquare, accumulateProduct, accumulateWeighted

Python prototype (for reference only):

accumulate(src, dst[, mask]) -> dst
Link to this function

accumulateProduct(src1, src2, dst)

View Source

Adds the per-element product of two input images to the accumulator image.

Positional Arguments
  • src1: Evision.Mat.t().

    First input image, 1- or 3-channel, 8-bit or 32-bit floating point.

  • src2: Evision.Mat.t().

    Second input image of the same type and the same size as src1 .

Keyword Arguments
  • mask: Evision.Mat.t().

    Optional operation mask.

Return
  • dst: Evision.Mat.t().

    %Accumulator image with the same number of channels as input images, 32-bit or 64-bit floating-point.

The function adds the product of two images or their selected regions to the accumulator dst : \f[\texttt{dst} (x,y) \leftarrow \texttt{dst} (x,y) + \texttt{src1} (x,y) \cdot \texttt{src2} (x,y) \quad \text{if} \quad \texttt{mask} (x,y) \ne 0\f] The function supports multi-channel images. Each channel is processed independently.

@sa accumulate, accumulateSquare, accumulateWeighted

Python prototype (for reference only):

accumulateProduct(src1, src2, dst[, mask]) -> dst
Link to this function

accumulateProduct(src1, src2, dst, opts)

View Source
@spec accumulateProduct(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Adds the per-element product of two input images to the accumulator image.

Positional Arguments
  • src1: Evision.Mat.t().

    First input image, 1- or 3-channel, 8-bit or 32-bit floating point.

  • src2: Evision.Mat.t().

    Second input image of the same type and the same size as src1 .

Keyword Arguments
  • mask: Evision.Mat.t().

    Optional operation mask.

Return
  • dst: Evision.Mat.t().

    %Accumulator image with the same number of channels as input images, 32-bit or 64-bit floating-point.

The function adds the product of two images or their selected regions to the accumulator dst : \f[\texttt{dst} (x,y) \leftarrow \texttt{dst} (x,y) + \texttt{src1} (x,y) \cdot \texttt{src2} (x,y) \quad \text{if} \quad \texttt{mask} (x,y) \ne 0\f] The function supports multi-channel images. Each channel is processed independently.

@sa accumulate, accumulateSquare, accumulateWeighted

Python prototype (for reference only):

accumulateProduct(src1, src2, dst[, mask]) -> dst
Link to this function

accumulateSquare(src, dst)

View Source
@spec accumulateSquare(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

Adds the square of a source image to the accumulator image.

Positional Arguments
  • src: Evision.Mat.t().

    Input image as 1- or 3-channel, 8-bit or 32-bit floating point.

Keyword Arguments
  • mask: Evision.Mat.t().

    Optional operation mask.

Return
  • dst: Evision.Mat.t().

    %Accumulator image with the same number of channels as input image, 32-bit or 64-bit floating-point.

The function adds the input image src or its selected region, raised to a power of 2, to the accumulator dst : \f[\texttt{dst} (x,y) \leftarrow \texttt{dst} (x,y) + \texttt{src} (x,y)^2 \quad \text{if} \quad \texttt{mask} (x,y) \ne 0\f] The function supports multi-channel images. Each channel is processed independently.

@sa accumulateSquare, accumulateProduct, accumulateWeighted

Python prototype (for reference only):

accumulateSquare(src, dst[, mask]) -> dst
Link to this function

accumulateSquare(src, dst, opts)

View Source
@spec accumulateSquare(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Adds the square of a source image to the accumulator image.

Positional Arguments
  • src: Evision.Mat.t().

    Input image as 1- or 3-channel, 8-bit or 32-bit floating point.

Keyword Arguments
  • mask: Evision.Mat.t().

    Optional operation mask.

Return
  • dst: Evision.Mat.t().

    %Accumulator image with the same number of channels as input image, 32-bit or 64-bit floating-point.

The function adds the input image src or its selected region, raised to a power of 2, to the accumulator dst : \f[\texttt{dst} (x,y) \leftarrow \texttt{dst} (x,y) + \texttt{src} (x,y)^2 \quad \text{if} \quad \texttt{mask} (x,y) \ne 0\f] The function supports multi-channel images. Each channel is processed independently.

@sa accumulateSquare, accumulateProduct, accumulateWeighted

Python prototype (for reference only):

accumulateSquare(src, dst[, mask]) -> dst
Link to this function

accumulateWeighted(src, dst, alpha)

View Source
@spec accumulateWeighted(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  number()
) ::
  Evision.Mat.t() | {:error, String.t()}

Updates a running average.

Positional Arguments
  • src: Evision.Mat.t().

    Input image as 1- or 3-channel, 8-bit or 32-bit floating point.

  • alpha: double.

    Weight of the input image.

Keyword Arguments
  • mask: Evision.Mat.t().

    Optional operation mask.

Return
  • dst: Evision.Mat.t().

    %Accumulator image with the same number of channels as input image, 32-bit or 64-bit floating-point.

The function calculates the weighted sum of the input image src and the accumulator dst so that dst becomes a running average of a frame sequence: \f[\texttt{dst} (x,y) \leftarrow (1- \texttt{alpha} ) \cdot \texttt{dst} (x,y) + \texttt{alpha} \cdot \texttt{src} (x,y) \quad \text{if} \quad \texttt{mask} (x,y) \ne 0\f] That is, alpha regulates the update speed (how fast the accumulator "forgets" about earlier images). The function supports multi-channel images. Each channel is processed independently.

@sa accumulate, accumulateSquare, accumulateProduct

Python prototype (for reference only):

accumulateWeighted(src, dst, alpha[, mask]) -> dst
Link to this function

accumulateWeighted(src, dst, alpha, opts)

View Source
@spec accumulateWeighted(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  number(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Updates a running average.

Positional Arguments
  • src: Evision.Mat.t().

    Input image as 1- or 3-channel, 8-bit or 32-bit floating point.

  • alpha: double.

    Weight of the input image.

Keyword Arguments
  • mask: Evision.Mat.t().

    Optional operation mask.

Return
  • dst: Evision.Mat.t().

    %Accumulator image with the same number of channels as input image, 32-bit or 64-bit floating-point.

The function calculates the weighted sum of the input image src and the accumulator dst so that dst becomes a running average of a frame sequence: \f[\texttt{dst} (x,y) \leftarrow (1- \texttt{alpha} ) \cdot \texttt{dst} (x,y) + \texttt{alpha} \cdot \texttt{src} (x,y) \quad \text{if} \quad \texttt{mask} (x,y) \ne 0\f] That is, alpha regulates the update speed (how fast the accumulator "forgets" about earlier images). The function supports multi-channel images. Each channel is processed independently.

@sa accumulate, accumulateSquare, accumulateProduct

Python prototype (for reference only):

accumulateWeighted(src, dst, alpha[, mask]) -> dst
Link to this function

adaptiveThreshold(src, maxValue, adaptiveMethod, thresholdType, blockSize, c)

View Source
@spec adaptiveThreshold(
  Evision.Mat.maybe_mat_in(),
  number(),
  integer(),
  integer(),
  integer(),
  number()
) ::
  Evision.Mat.t() | {:error, String.t()}

Applies an adaptive threshold to an array.

Positional Arguments
  • src: Evision.Mat.t().

    Source 8-bit single-channel image.

  • maxValue: double.

    Non-zero value assigned to the pixels for which the condition is satisfied

  • adaptiveMethod: int.

    Adaptive thresholding algorithm to use, see #AdaptiveThresholdTypes. The #BORDER_REPLICATE | #BORDER_ISOLATED is used to process boundaries.

  • thresholdType: int.

    Thresholding type that must be either #THRESH_BINARY or #THRESH_BINARY_INV, see #ThresholdTypes.

  • blockSize: int.

    Size of a pixel neighborhood that is used to calculate a threshold value for the pixel: 3, 5, 7, and so on.

  • c: double.

    Constant subtracted from the mean or weighted mean (see the details below). Normally, it is positive but may be zero or negative as well.

Return
  • dst: Evision.Mat.t().

    Destination image of the same size and the same type as src.

The function transforms a grayscale image to a binary image according to the formulae:

  • THRESH_BINARY \f[dst(x,y) = \fork{\texttt{maxValue}}{if (src(x,y) > T(x,y))}{0}{otherwise}\f]

  • THRESH_BINARY_INV \f[dst(x,y) = \fork{0}{if (src(x,y) > T(x,y))}{\texttt{maxValue}}{otherwise}\f] where \f$T(x,y)\f$ is a threshold calculated individually for each pixel (see adaptiveMethod parameter).

The function can process the image in-place.

@sa threshold, blur, GaussianBlur

Python prototype (for reference only):

adaptiveThreshold(src, maxValue, adaptiveMethod, thresholdType, blockSize, C[, dst]) -> dst
Link to this function

adaptiveThreshold(src, maxValue, adaptiveMethod, thresholdType, blockSize, c, opts)

View Source
@spec adaptiveThreshold(
  Evision.Mat.maybe_mat_in(),
  number(),
  integer(),
  integer(),
  integer(),
  number(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Applies an adaptive threshold to an array.

Positional Arguments
  • src: Evision.Mat.t().

    Source 8-bit single-channel image.

  • maxValue: double.

    Non-zero value assigned to the pixels for which the condition is satisfied

  • adaptiveMethod: int.

    Adaptive thresholding algorithm to use, see #AdaptiveThresholdTypes. The #BORDER_REPLICATE | #BORDER_ISOLATED is used to process boundaries.

  • thresholdType: int.

    Thresholding type that must be either #THRESH_BINARY or #THRESH_BINARY_INV, see #ThresholdTypes.

  • blockSize: int.

    Size of a pixel neighborhood that is used to calculate a threshold value for the pixel: 3, 5, 7, and so on.

  • c: double.

    Constant subtracted from the mean or weighted mean (see the details below). Normally, it is positive but may be zero or negative as well.

Return
  • dst: Evision.Mat.t().

    Destination image of the same size and the same type as src.

The function transforms a grayscale image to a binary image according to the formulae:

  • THRESH_BINARY \f[dst(x,y) = \fork{\texttt{maxValue}}{if (src(x,y) > T(x,y))}{0}{otherwise}\f]

  • THRESH_BINARY_INV \f[dst(x,y) = \fork{0}{if (src(x,y) > T(x,y))}{\texttt{maxValue}}{otherwise}\f] where \f$T(x,y)\f$ is a threshold calculated individually for each pixel (see adaptiveMethod parameter).

The function can process the image in-place.

@sa threshold, blur, GaussianBlur

Python prototype (for reference only):

adaptiveThreshold(src, maxValue, adaptiveMethod, thresholdType, blockSize, C[, dst]) -> dst

Calculates the per-element sum of two arrays or an array and a scalar.

Positional Arguments
  • src1: Evision.Mat.t().

    first input array or a scalar.

  • src2: Evision.Mat.t().

    second input array or a scalar.

Keyword Arguments
  • mask: Evision.Mat.t().

    optional operation mask - 8-bit single channel array, that specifies elements of the output array to be changed.

  • dtype: int.

    optional depth of the output array (see the discussion below).

Return
  • dst: Evision.Mat.t().

    output array that has the same size and number of channels as the input array(s); the depth is defined by dtype or src1/src2.

The function add calculates:

  • Sum of two arrays when both input arrays have the same size and the same number of channels: \f[\texttt{dst}(I) = \texttt{saturate} ( \texttt{src1}(I) + \texttt{src2}(I)) \quad \texttt{if mask}(I) \ne0\f]

  • Sum of an array and a scalar when src2 is constructed from Scalar or has the same number of elements as src1.channels(): \f[\texttt{dst}(I) = \texttt{saturate} ( \texttt{src1}(I) + \texttt{src2} ) \quad \texttt{if mask}(I) \ne0\f]

  • Sum of a scalar and an array when src1 is constructed from Scalar or has the same number of elements as src2.channels(): \f[\texttt{dst}(I) = \texttt{saturate} ( \texttt{src1} + \texttt{src2}(I) ) \quad \texttt{if mask}(I) \ne0\f] where I is a multi-dimensional index of array elements. In case of multi-channel arrays, each channel is processed independently.

The first function in the list above can be replaced with matrix expressions:

dst = src1 + src2;
dst += src1; // equivalent to add(dst, src1, dst);

The input arrays and the output array can all have the same or different depths. For example, you can add a 16-bit unsigned array to a 8-bit signed array and store the sum as a 32-bit floating-point array. Depth of the output array is determined by the dtype parameter. In the second and third cases above, as well as in the first case, when src1.depth() == src2.depth(), dtype can be set to the default -1. In this case, the output array will have the same depth as the input array, be it src1, src2 or both. Note: Saturation is not applied when the output array has the depth CV_32S. You may even get result of an incorrect sign in the case of overflow. Note: (Python) Be careful to difference behaviour between src1/src2 are single number and they are tuple/array. add(src,X) means add(src,(X,X,X,X)). add(src,(X,)) means add(src,(X,0,0,0)). @sa subtract, addWeighted, scaleAdd, Mat::convertTo

Python prototype (for reference only):

add(src1, src2[, dst[, mask[, dtype]]]) -> dst
@spec add(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Calculates the per-element sum of two arrays or an array and a scalar.

Positional Arguments
  • src1: Evision.Mat.t().

    first input array or a scalar.

  • src2: Evision.Mat.t().

    second input array or a scalar.

Keyword Arguments
  • mask: Evision.Mat.t().

    optional operation mask - 8-bit single channel array, that specifies elements of the output array to be changed.

  • dtype: int.

    optional depth of the output array (see the discussion below).

Return
  • dst: Evision.Mat.t().

    output array that has the same size and number of channels as the input array(s); the depth is defined by dtype or src1/src2.

The function add calculates:

  • Sum of two arrays when both input arrays have the same size and the same number of channels: \f[\texttt{dst}(I) = \texttt{saturate} ( \texttt{src1}(I) + \texttt{src2}(I)) \quad \texttt{if mask}(I) \ne0\f]

  • Sum of an array and a scalar when src2 is constructed from Scalar or has the same number of elements as src1.channels(): \f[\texttt{dst}(I) = \texttt{saturate} ( \texttt{src1}(I) + \texttt{src2} ) \quad \texttt{if mask}(I) \ne0\f]

  • Sum of a scalar and an array when src1 is constructed from Scalar or has the same number of elements as src2.channels(): \f[\texttt{dst}(I) = \texttt{saturate} ( \texttt{src1} + \texttt{src2}(I) ) \quad \texttt{if mask}(I) \ne0\f] where I is a multi-dimensional index of array elements. In case of multi-channel arrays, each channel is processed independently.

The first function in the list above can be replaced with matrix expressions:

dst = src1 + src2;
dst += src1; // equivalent to add(dst, src1, dst);

The input arrays and the output array can all have the same or different depths. For example, you can add a 16-bit unsigned array to a 8-bit signed array and store the sum as a 32-bit floating-point array. Depth of the output array is determined by the dtype parameter. In the second and third cases above, as well as in the first case, when src1.depth() == src2.depth(), dtype can be set to the default -1. In this case, the output array will have the same depth as the input array, be it src1, src2 or both. Note: Saturation is not applied when the output array has the depth CV_32S. You may even get result of an incorrect sign in the case of overflow. Note: (Python) Be careful to difference behaviour between src1/src2 are single number and they are tuple/array. add(src,X) means add(src,(X,X,X,X)). add(src,(X,)) means add(src,(X,0,0,0)). @sa subtract, addWeighted, scaleAdd, Mat::convertTo

Python prototype (for reference only):

add(src1, src2[, dst[, mask[, dtype]]]) -> dst
Link to this function

addText(img, text, org, nameFont)

View Source
@spec addText(Evision.Mat.maybe_mat_in(), binary(), {number(), number()}, binary()) ::
  :ok | {:error, String.t()}

Draws a text on the image.

Positional Arguments
  • img: Evision.Mat.t().

    8-bit 3-channel image where the text should be drawn.

  • text: String.

    Text to write on an image.

  • org: Point.

    Point(x,y) where the text should start on an image.

  • nameFont: String.

    Name of the font. The name should match the name of a system font (such as Times*). If the font is not found, a default one is used.

Keyword Arguments
  • pointSize: int.

    Size of the font. If not specified, equal zero or negative, the point size of the font is set to a system-dependent default value. Generally, this is 12 points.

  • color: Scalar.

    Color of the font in BGRA where A = 255 is fully transparent.

  • weight: int.

    Font weight. Available operation flags are : cv::QtFontWeights You can also specify a positive integer for better control.

  • style: int.

    Font style. Available operation flags are : cv::QtFontStyles

  • spacing: int.

    Spacing between characters. It can be negative or positive.

Python prototype (for reference only):

addText(img, text, org, nameFont[, pointSize[, color[, weight[, style[, spacing]]]]]) -> None
Link to this function

addText(img, text, org, nameFont, opts)

View Source
@spec addText(
  Evision.Mat.maybe_mat_in(),
  binary(),
  {number(), number()},
  binary(),
  [{atom(), term()}, ...] | nil
) :: :ok | {:error, String.t()}

Draws a text on the image.

Positional Arguments
  • img: Evision.Mat.t().

    8-bit 3-channel image where the text should be drawn.

  • text: String.

    Text to write on an image.

  • org: Point.

    Point(x,y) where the text should start on an image.

  • nameFont: String.

    Name of the font. The name should match the name of a system font (such as Times*). If the font is not found, a default one is used.

Keyword Arguments
  • pointSize: int.

    Size of the font. If not specified, equal zero or negative, the point size of the font is set to a system-dependent default value. Generally, this is 12 points.

  • color: Scalar.

    Color of the font in BGRA where A = 255 is fully transparent.

  • weight: int.

    Font weight. Available operation flags are : cv::QtFontWeights You can also specify a positive integer for better control.

  • style: int.

    Font style. Available operation flags are : cv::QtFontStyles

  • spacing: int.

    Spacing between characters. It can be negative or positive.

Python prototype (for reference only):

addText(img, text, org, nameFont[, pointSize[, color[, weight[, style[, spacing]]]]]) -> None
Link to this function

addWeighted(src1, alpha, src2, beta, gamma)

View Source
@spec addWeighted(
  Evision.Mat.maybe_mat_in(),
  number(),
  Evision.Mat.maybe_mat_in(),
  number(),
  number()
) ::
  Evision.Mat.t() | {:error, String.t()}

Calculates the weighted sum of two arrays.

Positional Arguments
  • src1: Evision.Mat.t().

    first input array.

  • alpha: double.

    weight of the first array elements.

  • src2: Evision.Mat.t().

    second input array of the same size and channel number as src1.

  • beta: double.

    weight of the second array elements.

  • gamma: double.

    scalar added to each sum.

Keyword Arguments
  • dtype: int.

    optional depth of the output array; when both input arrays have the same depth, dtype can be set to -1, which will be equivalent to src1.depth().

Return
  • dst: Evision.Mat.t().

    output array that has the same size and number of channels as the input arrays.

The function addWeighted calculates the weighted sum of two arrays as follows: \f[\texttt{dst} (I)= \texttt{saturate} ( \texttt{src1} (I)* \texttt{alpha} + \texttt{src2} (I)* \texttt{beta} + \texttt{gamma} )\f] where I is a multi-dimensional index of array elements. In case of multi-channel arrays, each channel is processed independently. The function can be replaced with a matrix expression:

dst = src1*alpha + src2*beta + gamma;

Note: Saturation is not applied when the output array has the depth CV_32S. You may even get result of an incorrect sign in the case of overflow. @sa add, subtract, scaleAdd, Mat::convertTo

Python prototype (for reference only):

addWeighted(src1, alpha, src2, beta, gamma[, dst[, dtype]]) -> dst
Link to this function

addWeighted(src1, alpha, src2, beta, gamma, opts)

View Source
@spec addWeighted(
  Evision.Mat.maybe_mat_in(),
  number(),
  Evision.Mat.maybe_mat_in(),
  number(),
  number(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Calculates the weighted sum of two arrays.

Positional Arguments
  • src1: Evision.Mat.t().

    first input array.

  • alpha: double.

    weight of the first array elements.

  • src2: Evision.Mat.t().

    second input array of the same size and channel number as src1.

  • beta: double.

    weight of the second array elements.

  • gamma: double.

    scalar added to each sum.

Keyword Arguments
  • dtype: int.

    optional depth of the output array; when both input arrays have the same depth, dtype can be set to -1, which will be equivalent to src1.depth().

Return
  • dst: Evision.Mat.t().

    output array that has the same size and number of channels as the input arrays.

The function addWeighted calculates the weighted sum of two arrays as follows: \f[\texttt{dst} (I)= \texttt{saturate} ( \texttt{src1} (I)* \texttt{alpha} + \texttt{src2} (I)* \texttt{beta} + \texttt{gamma} )\f] where I is a multi-dimensional index of array elements. In case of multi-channel arrays, each channel is processed independently. The function can be replaced with a matrix expression:

dst = src1*alpha + src2*beta + gamma;

Note: Saturation is not applied when the output array has the depth CV_32S. You may even get result of an incorrect sign in the case of overflow. @sa add, subtract, scaleAdd, Mat::convertTo

Python prototype (for reference only):

addWeighted(src1, alpha, src2, beta, gamma[, dst[, dtype]]) -> dst
Link to this function

applyColorMap(src, userColor)

View Source
@spec applyColorMap(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}
@spec applyColorMap(Evision.Mat.maybe_mat_in(), integer()) ::
  Evision.Mat.t() | {:error, String.t()}

Variant 1:

Applies a user colormap on a given image.

Positional Arguments
  • src: Evision.Mat.t().

    The source image, grayscale or colored of type CV_8UC1 or CV_8UC3.

  • userColor: Evision.Mat.t().

    The colormap to apply of type CV_8UC1 or CV_8UC3 and size 256

Return
  • dst: Evision.Mat.t().

    The result is the colormapped source image. Note: Mat::create is called on dst.

Python prototype (for reference only):

applyColorMap(src, userColor[, dst]) -> dst

Variant 2:

Applies a GNU Octave/MATLAB equivalent colormap on a given image.

Positional Arguments
  • src: Evision.Mat.t().

    The source image, grayscale or colored of type CV_8UC1 or CV_8UC3.

  • colormap: int.

    The colormap to apply, see #ColormapTypes

Return
  • dst: Evision.Mat.t().

    The result is the colormapped source image. Note: Mat::create is called on dst.

Python prototype (for reference only):

applyColorMap(src, colormap[, dst]) -> dst
Link to this function

applyColorMap(src, userColor, opts)

View Source
@spec applyColorMap(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}
@spec applyColorMap(
  Evision.Mat.maybe_mat_in(),
  integer(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Variant 1:

Applies a user colormap on a given image.

Positional Arguments
  • src: Evision.Mat.t().

    The source image, grayscale or colored of type CV_8UC1 or CV_8UC3.

  • userColor: Evision.Mat.t().

    The colormap to apply of type CV_8UC1 or CV_8UC3 and size 256

Return
  • dst: Evision.Mat.t().

    The result is the colormapped source image. Note: Mat::create is called on dst.

Python prototype (for reference only):

applyColorMap(src, userColor[, dst]) -> dst

Variant 2:

Applies a GNU Octave/MATLAB equivalent colormap on a given image.

Positional Arguments
  • src: Evision.Mat.t().

    The source image, grayscale or colored of type CV_8UC1 or CV_8UC3.

  • colormap: int.

    The colormap to apply, see #ColormapTypes

Return
  • dst: Evision.Mat.t().

    The result is the colormapped source image. Note: Mat::create is called on dst.

Python prototype (for reference only):

applyColorMap(src, colormap[, dst]) -> dst
Link to this function

approxPolyDP(curve, epsilon, closed)

View Source
@spec approxPolyDP(Evision.Mat.maybe_mat_in(), number(), boolean()) ::
  Evision.Mat.t() | {:error, String.t()}

Approximates a polygonal curve(s) with the specified precision.

Positional Arguments
  • curve: Evision.Mat.t().

    Input vector of a 2D point stored in std::vector or Mat

  • epsilon: double.

    Parameter specifying the approximation accuracy. This is the maximum distance between the original curve and its approximation.

  • closed: bool.

    If true, the approximated curve is closed (its first and last vertices are connected). Otherwise, it is not closed.

Return
  • approxCurve: Evision.Mat.t().

    Result of the approximation. The type should match the type of the input curve.

The function cv::approxPolyDP approximates a curve or a polygon with another curve/polygon with less vertices so that the distance between them is less or equal to the specified precision. It uses the Douglas-Peucker algorithm http://en.wikipedia.org/wiki/Ramer-Douglas-Peucker_algorithm

Python prototype (for reference only):

approxPolyDP(curve, epsilon, closed[, approxCurve]) -> approxCurve
Link to this function

approxPolyDP(curve, epsilon, closed, opts)

View Source
@spec approxPolyDP(
  Evision.Mat.maybe_mat_in(),
  number(),
  boolean(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Approximates a polygonal curve(s) with the specified precision.

Positional Arguments
  • curve: Evision.Mat.t().

    Input vector of a 2D point stored in std::vector or Mat

  • epsilon: double.

    Parameter specifying the approximation accuracy. This is the maximum distance between the original curve and its approximation.

  • closed: bool.

    If true, the approximated curve is closed (its first and last vertices are connected). Otherwise, it is not closed.

Return
  • approxCurve: Evision.Mat.t().

    Result of the approximation. The type should match the type of the input curve.

The function cv::approxPolyDP approximates a curve or a polygon with another curve/polygon with less vertices so that the distance between them is less or equal to the specified precision. It uses the Douglas-Peucker algorithm http://en.wikipedia.org/wiki/Ramer-Douglas-Peucker_algorithm

Python prototype (for reference only):

approxPolyDP(curve, epsilon, closed[, approxCurve]) -> approxCurve
Link to this function

arcLength(curve, closed)

View Source
@spec arcLength(Evision.Mat.maybe_mat_in(), boolean()) ::
  number() | {:error, String.t()}

Calculates a contour perimeter or a curve length.

Positional Arguments
  • curve: Evision.Mat.t().

    Input vector of 2D points, stored in std::vector or Mat.

  • closed: bool.

    Flag indicating whether the curve is closed or not.

Return
  • retval: double

The function computes a curve length or a closed contour perimeter.

Python prototype (for reference only):

arcLength(curve, closed) -> retval
Link to this function

arrowedLine(img, pt1, pt2, color)

View Source
@spec arrowedLine(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  {number(), number()},
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()}
) :: Evision.Mat.t() | {:error, String.t()}

Draws an arrow segment pointing from the first point to the second one.

Positional Arguments
  • pt1: Point.

    The point the arrow starts from.

  • pt2: Point.

    The point the arrow points to.

  • color: Scalar.

    Line color.

Keyword Arguments
  • thickness: int.

    Line thickness.

  • line_type: int.

    Type of the line. See #LineTypes

  • shift: int.

    Number of fractional bits in the point coordinates.

  • tipLength: double.

    The length of the arrow tip in relation to the arrow length

Return
  • img: Evision.Mat.t().

    Image.

The function cv::arrowedLine draws an arrow between pt1 and pt2 points in the image. See also #line.

Python prototype (for reference only):

arrowedLine(img, pt1, pt2, color[, thickness[, line_type[, shift[, tipLength]]]]) -> img
Link to this function

arrowedLine(img, pt1, pt2, color, opts)

View Source
@spec arrowedLine(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  {number(), number()},
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()},
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Draws an arrow segment pointing from the first point to the second one.

Positional Arguments
  • pt1: Point.

    The point the arrow starts from.

  • pt2: Point.

    The point the arrow points to.

  • color: Scalar.

    Line color.

Keyword Arguments
  • thickness: int.

    Line thickness.

  • line_type: int.

    Type of the line. See #LineTypes

  • shift: int.

    Number of fractional bits in the point coordinates.

  • tipLength: double.

    The length of the arrow tip in relation to the arrow length

Return
  • img: Evision.Mat.t().

    Image.

The function cv::arrowedLine draws an arrow between pt1 and pt2 points in the image. See also #line.

Python prototype (for reference only):

arrowedLine(img, pt1, pt2, color[, thickness[, line_type[, shift[, tipLength]]]]) -> img
Link to this function

batchDistance(src1, src2, dtype)

View Source
@spec batchDistance(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in(), integer()) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

naive nearest neighbor finder

Positional Arguments
  • src1: Evision.Mat.t()
  • src2: Evision.Mat.t()
  • dtype: int
Keyword Arguments
  • normType: int.
  • k: int.
  • mask: Evision.Mat.t().
  • update: int.
  • crosscheck: bool.
Return
  • dist: Evision.Mat.t().
  • nidx: Evision.Mat.t().

see http://en.wikipedia.org/wiki/Nearest_neighbor_search @todo document

Python prototype (for reference only):

batchDistance(src1, src2, dtype[, dist[, nidx[, normType[, K[, mask[, update[, crosscheck]]]]]]]) -> dist, nidx
Link to this function

batchDistance(src1, src2, dtype, opts)

View Source
@spec batchDistance(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  integer(),
  [{atom(), term()}, ...] | nil
) :: {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

naive nearest neighbor finder

Positional Arguments
  • src1: Evision.Mat.t()
  • src2: Evision.Mat.t()
  • dtype: int
Keyword Arguments
  • normType: int.
  • k: int.
  • mask: Evision.Mat.t().
  • update: int.
  • crosscheck: bool.
Return
  • dist: Evision.Mat.t().
  • nidx: Evision.Mat.t().

see http://en.wikipedia.org/wiki/Nearest_neighbor_search @todo document

Python prototype (for reference only):

batchDistance(src1, src2, dtype[, dist[, nidx[, normType[, K[, mask[, update[, crosscheck]]]]]]]) -> dist, nidx
Link to this function

bilateralFilter(src, d, sigmaColor, sigmaSpace)

View Source
@spec bilateralFilter(Evision.Mat.maybe_mat_in(), integer(), number(), number()) ::
  Evision.Mat.t() | {:error, String.t()}

Applies the bilateral filter to an image.

Positional Arguments
  • src: Evision.Mat.t().

    Source 8-bit or floating-point, 1-channel or 3-channel image.

  • d: int.

    Diameter of each pixel neighborhood that is used during filtering. If it is non-positive, it is computed from sigmaSpace.

  • sigmaColor: double.

    Filter sigma in the color space. A larger value of the parameter means that farther colors within the pixel neighborhood (see sigmaSpace) will be mixed together, resulting in larger areas of semi-equal color.

  • sigmaSpace: double.

    Filter sigma in the coordinate space. A larger value of the parameter means that farther pixels will influence each other as long as their colors are close enough (see sigmaColor ). When d>0, it specifies the neighborhood size regardless of sigmaSpace. Otherwise, d is proportional to sigmaSpace.

Keyword Arguments
  • borderType: int.

    border mode used to extrapolate pixels outside of the image, see #BorderTypes

Return
  • dst: Evision.Mat.t().

    Destination image of the same size and type as src .

The function applies bilateral filtering to the input image, as described in http://www.dai.ed.ac.uk/CVonline/LOCAL_COPIES/MANDUCHI1/Bilateral_Filtering.html bilateralFilter can reduce unwanted noise very well while keeping edges fairly sharp. However, it is very slow compared to most filters. Sigma values: For simplicity, you can set the 2 sigma values to be the same. If they are small (\< 10), the filter will not have much effect, whereas if they are large (> 150), they will have a very strong effect, making the image look "cartoonish". Filter size: Large filters (d > 5) are very slow, so it is recommended to use d=5 for real-time applications, and perhaps d=9 for offline applications that need heavy noise filtering. This filter does not work inplace.

Python prototype (for reference only):

bilateralFilter(src, d, sigmaColor, sigmaSpace[, dst[, borderType]]) -> dst
Link to this function

bilateralFilter(src, d, sigmaColor, sigmaSpace, opts)

View Source
@spec bilateralFilter(
  Evision.Mat.maybe_mat_in(),
  integer(),
  number(),
  number(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Applies the bilateral filter to an image.

Positional Arguments
  • src: Evision.Mat.t().

    Source 8-bit or floating-point, 1-channel or 3-channel image.

  • d: int.

    Diameter of each pixel neighborhood that is used during filtering. If it is non-positive, it is computed from sigmaSpace.

  • sigmaColor: double.

    Filter sigma in the color space. A larger value of the parameter means that farther colors within the pixel neighborhood (see sigmaSpace) will be mixed together, resulting in larger areas of semi-equal color.

  • sigmaSpace: double.

    Filter sigma in the coordinate space. A larger value of the parameter means that farther pixels will influence each other as long as their colors are close enough (see sigmaColor ). When d>0, it specifies the neighborhood size regardless of sigmaSpace. Otherwise, d is proportional to sigmaSpace.

Keyword Arguments
  • borderType: int.

    border mode used to extrapolate pixels outside of the image, see #BorderTypes

Return
  • dst: Evision.Mat.t().

    Destination image of the same size and type as src .

The function applies bilateral filtering to the input image, as described in http://www.dai.ed.ac.uk/CVonline/LOCAL_COPIES/MANDUCHI1/Bilateral_Filtering.html bilateralFilter can reduce unwanted noise very well while keeping edges fairly sharp. However, it is very slow compared to most filters. Sigma values: For simplicity, you can set the 2 sigma values to be the same. If they are small (\< 10), the filter will not have much effect, whereas if they are large (> 150), they will have a very strong effect, making the image look "cartoonish". Filter size: Large filters (d > 5) are very slow, so it is recommended to use d=5 for real-time applications, and perhaps d=9 for offline applications that need heavy noise filtering. This filter does not work inplace.

Python prototype (for reference only):

bilateralFilter(src, d, sigmaColor, sigmaSpace[, dst[, borderType]]) -> dst
Link to this function

blendLinear(src1, src2, weights1, weights2)

View Source

blendLinear

Positional Arguments
  • src1: Evision.Mat.t()
  • src2: Evision.Mat.t()
  • weights1: Evision.Mat.t()
  • weights2: Evision.Mat.t()
Return
  • dst: Evision.Mat.t().

Has overloading in C++

variant without mask parameter

Python prototype (for reference only):

blendLinear(src1, src2, weights1, weights2[, dst]) -> dst
Link to this function

blendLinear(src1, src2, weights1, weights2, opts)

View Source

blendLinear

Positional Arguments
  • src1: Evision.Mat.t()
  • src2: Evision.Mat.t()
  • weights1: Evision.Mat.t()
  • weights2: Evision.Mat.t()
Return
  • dst: Evision.Mat.t().

Has overloading in C++

variant without mask parameter

Python prototype (for reference only):

blendLinear(src1, src2, weights1, weights2[, dst]) -> dst
@spec blur(
  Evision.Mat.maybe_mat_in(),
  {number(), number()}
) :: Evision.Mat.t() | {:error, String.t()}

Blurs an image using the normalized box filter.

Positional Arguments
  • src: Evision.Mat.t().

    input image; it can have any number of channels, which are processed independently, but the depth should be CV_8U, CV_16U, CV_16S, CV_32F or CV_64F.

  • ksize: Size.

    blurring kernel size.

Keyword Arguments
  • anchor: Point.

    anchor point; default value Point(-1,-1) means that the anchor is at the kernel center.

  • borderType: int.

    border mode used to extrapolate pixels outside of the image, see #BorderTypes. #BORDER_WRAP is not supported.

Return
  • dst: Evision.Mat.t().

    output image of the same size and type as src.

The function smooths an image using the kernel: \f[\texttt{K} = \frac{1}{\texttt{ksize.width*ksize.height}} \begin{bmatrix} 1 & 1 & 1 & \cdots & 1 & 1 \\ 1 & 1 & 1 & \cdots & 1 & 1 \\ \hdotsfor{6} \\ 1 & 1 & 1 & \cdots & 1 & 1 \\ \end{bmatrix}\f] The call blur(src, dst, ksize, anchor, borderType) is equivalent to boxFilter(src, dst, src.type(), ksize, anchor, true, borderType). @sa boxFilter, bilateralFilter, GaussianBlur, medianBlur

Python prototype (for reference only):

blur(src, ksize[, dst[, anchor[, borderType]]]) -> dst
@spec blur(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Blurs an image using the normalized box filter.

Positional Arguments
  • src: Evision.Mat.t().

    input image; it can have any number of channels, which are processed independently, but the depth should be CV_8U, CV_16U, CV_16S, CV_32F or CV_64F.

  • ksize: Size.

    blurring kernel size.

Keyword Arguments
  • anchor: Point.

    anchor point; default value Point(-1,-1) means that the anchor is at the kernel center.

  • borderType: int.

    border mode used to extrapolate pixels outside of the image, see #BorderTypes. #BORDER_WRAP is not supported.

Return
  • dst: Evision.Mat.t().

    output image of the same size and type as src.

The function smooths an image using the kernel: \f[\texttt{K} = \frac{1}{\texttt{ksize.width*ksize.height}} \begin{bmatrix} 1 & 1 & 1 & \cdots & 1 & 1 \\ 1 & 1 & 1 & \cdots & 1 & 1 \\ \hdotsfor{6} \\ 1 & 1 & 1 & \cdots & 1 & 1 \\ \end{bmatrix}\f] The call blur(src, dst, ksize, anchor, borderType) is equivalent to boxFilter(src, dst, src.type(), ksize, anchor, true, borderType). @sa boxFilter, bilateralFilter, GaussianBlur, medianBlur

Python prototype (for reference only):

blur(src, ksize[, dst[, anchor[, borderType]]]) -> dst
Link to this function

borderInterpolate(p, len, borderType)

View Source
@spec borderInterpolate(integer(), integer(), integer()) ::
  integer() | {:error, String.t()}

Computes the source location of an extrapolated pixel.

Positional Arguments
  • p: int.

    0-based coordinate of the extrapolated pixel along one of the axes, likely \<0 or >= len

  • len: int.

    Length of the array along the corresponding axis.

  • borderType: int.

    Border type, one of the #BorderTypes, except for #BORDER_TRANSPARENT and #BORDER_ISOLATED . When borderType==#BORDER_CONSTANT , the function always returns -1, regardless of p and len.

Return
  • retval: int

The function computes and returns the coordinate of a donor pixel corresponding to the specified extrapolated pixel when using the specified extrapolation border mode. For example, if you use cv::BORDER_WRAP mode in the horizontal direction, cv::BORDER_REFLECT_101 in the vertical direction and want to compute value of the "virtual" pixel Point(-5, 100) in a floating-point image img , it looks like:

float val = img.at<float>(borderInterpolate(100, img.rows, cv::BORDER_REFLECT_101),
borderInterpolate(-5, img.cols, cv::BORDER_WRAP));

Normally, the function is not called directly. It is used inside filtering functions and also in copyMakeBorder.

@sa copyMakeBorder

Python prototype (for reference only):

borderInterpolate(p, len, borderType) -> retval
@spec boundingRect(Evision.Mat.maybe_mat_in()) ::
  {number(), number(), number(), number()} | {:error, String.t()}

Calculates the up-right bounding rectangle of a point set or non-zero pixels of gray-scale image.

Positional Arguments
  • array: Evision.Mat.t().

    Input gray-scale image or 2D point set, stored in std::vector or Mat.

Return
  • retval: Rect

The function calculates and returns the minimal up-right bounding rectangle for the specified point set or non-zero pixels of gray-scale image.

Python prototype (for reference only):

boundingRect(array) -> retval
Link to this function

boxFilter(src, ddepth, ksize)

View Source
@spec boxFilter(Evision.Mat.maybe_mat_in(), integer(), {number(), number()}) ::
  Evision.Mat.t() | {:error, String.t()}

Blurs an image using the box filter.

Positional Arguments
  • src: Evision.Mat.t().

    input image.

  • ddepth: int.

    the output image depth (-1 to use src.depth()).

  • ksize: Size.

    blurring kernel size.

Keyword Arguments
  • anchor: Point.

    anchor point; default value Point(-1,-1) means that the anchor is at the kernel center.

  • normalize: bool.

    flag, specifying whether the kernel is normalized by its area or not.

  • borderType: int.

    border mode used to extrapolate pixels outside of the image, see #BorderTypes. #BORDER_WRAP is not supported.

Return
  • dst: Evision.Mat.t().

    output image of the same size and type as src.

The function smooths an image using the kernel: \f[\texttt{K} = \alpha \begin{bmatrix} 1 & 1 & 1 & \cdots & 1 & 1 \\ 1 & 1 & 1 & \cdots & 1 & 1 \\ \hdotsfor{6} \\ 1 & 1 & 1 & \cdots & 1 & 1 \end{bmatrix}\f] where \f[\alpha = \begin{cases} \frac{1}{\texttt{ksize.width*ksize.height}} & \texttt{when } \texttt{normalize=true} \\1 & \texttt{otherwise}\end{cases}\f] Unnormalized box filter is useful for computing various integral characteristics over each pixel neighborhood, such as covariance matrices of image derivatives (used in dense optical flow algorithms, and so on). If you need to compute pixel sums over variable-size windows, use #integral. @sa blur, bilateralFilter, GaussianBlur, medianBlur, integral

Python prototype (for reference only):

boxFilter(src, ddepth, ksize[, dst[, anchor[, normalize[, borderType]]]]) -> dst
Link to this function

boxFilter(src, ddepth, ksize, opts)

View Source
@spec boxFilter(
  Evision.Mat.maybe_mat_in(),
  integer(),
  {number(), number()},
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Blurs an image using the box filter.

Positional Arguments
  • src: Evision.Mat.t().

    input image.

  • ddepth: int.

    the output image depth (-1 to use src.depth()).

  • ksize: Size.

    blurring kernel size.

Keyword Arguments
  • anchor: Point.

    anchor point; default value Point(-1,-1) means that the anchor is at the kernel center.

  • normalize: bool.

    flag, specifying whether the kernel is normalized by its area or not.

  • borderType: int.

    border mode used to extrapolate pixels outside of the image, see #BorderTypes. #BORDER_WRAP is not supported.

Return
  • dst: Evision.Mat.t().

    output image of the same size and type as src.

The function smooths an image using the kernel: \f[\texttt{K} = \alpha \begin{bmatrix} 1 & 1 & 1 & \cdots & 1 & 1 \\ 1 & 1 & 1 & \cdots & 1 & 1 \\ \hdotsfor{6} \\ 1 & 1 & 1 & \cdots & 1 & 1 \end{bmatrix}\f] where \f[\alpha = \begin{cases} \frac{1}{\texttt{ksize.width*ksize.height}} & \texttt{when } \texttt{normalize=true} \\1 & \texttt{otherwise}\end{cases}\f] Unnormalized box filter is useful for computing various integral characteristics over each pixel neighborhood, such as covariance matrices of image derivatives (used in dense optical flow algorithms, and so on). If you need to compute pixel sums over variable-size windows, use #integral. @sa blur, bilateralFilter, GaussianBlur, medianBlur, integral

Python prototype (for reference only):

boxFilter(src, ddepth, ksize[, dst[, anchor[, normalize[, borderType]]]]) -> dst
@spec boxPoints(Evision.RotatedRect.t()) :: Evision.Mat.t() | {:error, String.t()}

Finds the four vertices of a rotated rect. Useful to draw the rotated rectangle.

Positional Arguments
  • box: {centre={x, y}, size={s1, s2}, angle}.

    The input rotated rectangle. It may be the output of @ref minAreaRect.

Return
  • points: Evision.Mat.t().

    The output array of four vertices of rectangles.

The function finds the four vertices of a rotated rectangle. This function is useful to draw the rectangle. In C++, instead of using this function, you can directly use RotatedRect::points method. Please visit the @ref tutorial_bounding_rotated_ellipses "tutorial on Creating Bounding rotated boxes and ellipses for contours" for more information.

Python prototype (for reference only):

boxPoints(box[, points]) -> points
@spec boxPoints(Evision.RotatedRect.t(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Finds the four vertices of a rotated rect. Useful to draw the rotated rectangle.

Positional Arguments
  • box: {centre={x, y}, size={s1, s2}, angle}.

    The input rotated rectangle. It may be the output of @ref minAreaRect.

Return
  • points: Evision.Mat.t().

    The output array of four vertices of rectangles.

The function finds the four vertices of a rotated rectangle. This function is useful to draw the rectangle. In C++, instead of using this function, you can directly use RotatedRect::points method. Please visit the @ref tutorial_bounding_rotated_ellipses "tutorial on Creating Bounding rotated boxes and ellipses for contours" for more information.

Python prototype (for reference only):

boxPoints(box[, points]) -> points
@spec broadcast(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

Broadcast the given Mat to the given shape.

Positional Arguments
  • src: Evision.Mat.t().

    input array

  • shape: Evision.Mat.t().

    target shape. Should be a list of CV_32S numbers. Note that negative values are not supported.

Return
  • dst: Evision.Mat.t().

    output array that has the given shape

Python prototype (for reference only):

broadcast(src, shape[, dst]) -> dst
Link to this function

broadcast(src, shape, opts)

View Source
@spec broadcast(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Broadcast the given Mat to the given shape.

Positional Arguments
  • src: Evision.Mat.t().

    input array

  • shape: Evision.Mat.t().

    target shape. Should be a list of CV_32S numbers. Note that negative values are not supported.

Return
  • dst: Evision.Mat.t().

    output array that has the given shape

Python prototype (for reference only):

broadcast(src, shape[, dst]) -> dst
Link to this function

buildOpticalFlowPyramid(img, winSize, maxLevel)

View Source
@spec buildOpticalFlowPyramid(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  integer()
) ::
  {integer(), [Evision.Mat.t()]} | {:error, String.t()}

Constructs the image pyramid which can be passed to calcOpticalFlowPyrLK.

Positional Arguments
  • img: Evision.Mat.t().

    8-bit input image.

  • winSize: Size.

    window size of optical flow algorithm. Must be not less than winSize argument of calcOpticalFlowPyrLK. It is needed to calculate required padding for pyramid levels.

  • maxLevel: int.

    0-based maximal pyramid level number.

Keyword Arguments
  • withDerivatives: bool.

    set to precompute gradients for the every pyramid level. If pyramid is constructed without the gradients then calcOpticalFlowPyrLK will calculate them internally.

  • pyrBorder: int.

    the border mode for pyramid layers.

  • derivBorder: int.

    the border mode for gradients.

  • tryReuseInputImage: bool.

    put ROI of input image into the pyramid if possible. You can pass false to force data copying.

Return
  • retval: int

  • pyramid: [Evision.Mat].

    output pyramid.

@return number of levels in constructed pyramid. Can be less than maxLevel.

Python prototype (for reference only):

buildOpticalFlowPyramid(img, winSize, maxLevel[, pyramid[, withDerivatives[, pyrBorder[, derivBorder[, tryReuseInputImage]]]]]) -> retval, pyramid
Link to this function

buildOpticalFlowPyramid(img, winSize, maxLevel, opts)

View Source
@spec buildOpticalFlowPyramid(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  integer(),
  [{atom(), term()}, ...] | nil
) :: {integer(), [Evision.Mat.t()]} | {:error, String.t()}

Constructs the image pyramid which can be passed to calcOpticalFlowPyrLK.

Positional Arguments
  • img: Evision.Mat.t().

    8-bit input image.

  • winSize: Size.

    window size of optical flow algorithm. Must be not less than winSize argument of calcOpticalFlowPyrLK. It is needed to calculate required padding for pyramid levels.

  • maxLevel: int.

    0-based maximal pyramid level number.

Keyword Arguments
  • withDerivatives: bool.

    set to precompute gradients for the every pyramid level. If pyramid is constructed without the gradients then calcOpticalFlowPyrLK will calculate them internally.

  • pyrBorder: int.

    the border mode for pyramid layers.

  • derivBorder: int.

    the border mode for gradients.

  • tryReuseInputImage: bool.

    put ROI of input image into the pyramid if possible. You can pass false to force data copying.

Return
  • retval: int

  • pyramid: [Evision.Mat].

    output pyramid.

@return number of levels in constructed pyramid. Can be less than maxLevel.

Python prototype (for reference only):

buildOpticalFlowPyramid(img, winSize, maxLevel[, pyramid[, withDerivatives[, pyrBorder[, derivBorder[, tryReuseInputImage]]]]]) -> retval, pyramid
Link to this function

calcBackProject(images, channels, hist, ranges, scale)

View Source
@spec calcBackProject(
  [Evision.Mat.maybe_mat_in()],
  [integer()],
  Evision.Mat.maybe_mat_in(),
  [number()],
  number()
) :: Evision.Mat.t() | {:error, String.t()}

calcBackProject

Positional Arguments
  • images: [Evision.Mat]
  • channels: [int]
  • hist: Evision.Mat.t()
  • ranges: [float]
  • scale: double
Return
  • dst: Evision.Mat.t().

Has overloading in C++

Python prototype (for reference only):

calcBackProject(images, channels, hist, ranges, scale[, dst]) -> dst
Link to this function

calcBackProject(images, channels, hist, ranges, scale, opts)

View Source
@spec calcBackProject(
  [Evision.Mat.maybe_mat_in()],
  [integer()],
  Evision.Mat.maybe_mat_in(),
  [number()],
  number(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

calcBackProject

Positional Arguments
  • images: [Evision.Mat]
  • channels: [int]
  • hist: Evision.Mat.t()
  • ranges: [float]
  • scale: double
Return
  • dst: Evision.Mat.t().

Has overloading in C++

Python prototype (for reference only):

calcBackProject(images, channels, hist, ranges, scale[, dst]) -> dst
Link to this function

calcCovarMatrix(samples, mean, flags)

View Source
@spec calcCovarMatrix(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  integer()
) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

calcCovarMatrix

Positional Arguments
  • samples: Evision.Mat.t().

    samples stored as rows/columns of a single matrix.

  • flags: int.

    operation flags as a combination of #CovarFlags

Keyword Arguments
  • ctype: int.

    type of the matrixl; it equals 'CV_64F' by default.

Return
  • covar: Evision.Mat.t().

    output covariance matrix of the type ctype and square size.

  • mean: Evision.Mat.t().

    input or output (depending on the flags) array as the average value of the input vectors.

Has overloading in C++

Note: use #COVAR_ROWS or #COVAR_COLS flag

Python prototype (for reference only):

calcCovarMatrix(samples, mean, flags[, covar[, ctype]]) -> covar, mean
Link to this function

calcCovarMatrix(samples, mean, flags, opts)

View Source
@spec calcCovarMatrix(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  integer(),
  [{atom(), term()}, ...] | nil
) :: {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

calcCovarMatrix

Positional Arguments
  • samples: Evision.Mat.t().

    samples stored as rows/columns of a single matrix.

  • flags: int.

    operation flags as a combination of #CovarFlags

Keyword Arguments
  • ctype: int.

    type of the matrixl; it equals 'CV_64F' by default.

Return
  • covar: Evision.Mat.t().

    output covariance matrix of the type ctype and square size.

  • mean: Evision.Mat.t().

    input or output (depending on the flags) array as the average value of the input vectors.

Has overloading in C++

Note: use #COVAR_ROWS or #COVAR_COLS flag

Python prototype (for reference only):

calcCovarMatrix(samples, mean, flags[, covar[, ctype]]) -> covar, mean
Link to this function

calcHist(images, channels, mask, histSize, ranges)

View Source
@spec calcHist(
  [Evision.Mat.maybe_mat_in()],
  [integer()],
  Evision.Mat.maybe_mat_in(),
  [integer()],
  [
    number()
  ]
) :: Evision.Mat.t() | {:error, String.t()}

calcHist

Positional Arguments
  • images: [Evision.Mat]
  • channels: [int]
  • mask: Evision.Mat.t()
  • histSize: [int]
  • ranges: [float]
Keyword Arguments
  • accumulate: bool.
Return
  • hist: Evision.Mat.t().

Has overloading in C++

this variant supports only uniform histograms. ranges argument is either empty vector or a flattened vector of histSize.size()*2 elements (histSize.size() element pairs). The first and second elements of each pair specify the lower and upper boundaries.

Python prototype (for reference only):

calcHist(images, channels, mask, histSize, ranges[, hist[, accumulate]]) -> hist
Link to this function

calcHist(images, channels, mask, histSize, ranges, opts)

View Source
@spec calcHist(
  [Evision.Mat.maybe_mat_in()],
  [integer()],
  Evision.Mat.maybe_mat_in(),
  [integer()],
  [number()],
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

calcHist

Positional Arguments
  • images: [Evision.Mat]
  • channels: [int]
  • mask: Evision.Mat.t()
  • histSize: [int]
  • ranges: [float]
Keyword Arguments
  • accumulate: bool.
Return
  • hist: Evision.Mat.t().

Has overloading in C++

this variant supports only uniform histograms. ranges argument is either empty vector or a flattened vector of histSize.size()*2 elements (histSize.size() element pairs). The first and second elements of each pair specify the lower and upper boundaries.

Python prototype (for reference only):

calcHist(images, channels, mask, histSize, ranges[, hist[, accumulate]]) -> hist
Link to this function

calcOpticalFlowFarneback(prev, next, flow, pyr_scale, levels, winsize, iterations, poly_n, poly_sigma, flags)

View Source
@spec calcOpticalFlowFarneback(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  number(),
  integer(),
  integer(),
  integer(),
  integer(),
  number(),
  integer()
) :: Evision.Mat.t() | {:error, String.t()}

Computes a dense optical flow using the Gunnar Farneback's algorithm.

Positional Arguments
  • prev: Evision.Mat.t().

    first 8-bit single-channel input image.

  • next: Evision.Mat.t().

    second input image of the same size and the same type as prev.

  • pyr_scale: double.

    parameter, specifying the image scale (\<1) to build pyramids for each image; pyr_scale=0.5 means a classical pyramid, where each next layer is twice smaller than the previous one.

  • levels: int.

    number of pyramid layers including the initial image; levels=1 means that no extra layers are created and only the original images are used.

  • winsize: int.

    averaging window size; larger values increase the algorithm robustness to image noise and give more chances for fast motion detection, but yield more blurred motion field.

  • iterations: int.

    number of iterations the algorithm does at each pyramid level.

  • poly_n: int.

    size of the pixel neighborhood used to find polynomial expansion in each pixel; larger values mean that the image will be approximated with smoother surfaces, yielding more robust algorithm and more blurred motion field, typically poly_n =5 or 7.

  • poly_sigma: double.

    standard deviation of the Gaussian that is used to smooth derivatives used as a basis for the polynomial expansion; for poly_n=5, you can set poly_sigma=1.1, for poly_n=7, a good value would be poly_sigma=1.5.

  • flags: int.

    operation flags that can be a combination of the following:

    • OPTFLOW_USE_INITIAL_FLOW uses the input flow as an initial flow approximation.
    • OPTFLOW_FARNEBACK_GAUSSIAN uses the Gaussian \f$\texttt{winsize}\times\texttt{winsize}\f$ filter instead of a box filter of the same size for optical flow estimation; usually, this option gives z more accurate flow than with a box filter, at the cost of lower speed; normally, winsize for a Gaussian window should be set to a larger value to achieve the same level of robustness.
Return
  • flow: Evision.Mat.t().

    computed flow image that has the same size as prev and type CV_32FC2.

The function finds an optical flow for each prev pixel using the @cite Farneback2003 algorithm so that \f[\texttt{prev} (y,x) \sim \texttt{next} ( y + \texttt{flow} (y,x)[1], x + \texttt{flow} (y,x)[0])\f] Note: Some examples:

  • An example using the optical flow algorithm described by Gunnar Farneback can be found at opencv_source_code/samples/cpp/fback.cpp

  • (Python) An example using the optical flow algorithm described by Gunnar Farneback can be found at opencv_source_code/samples/python/opt_flow.py

Python prototype (for reference only):

calcOpticalFlowFarneback(prev, next, flow, pyr_scale, levels, winsize, iterations, poly_n, poly_sigma, flags) -> flow
Link to this function

calcOpticalFlowPyrLK(prevImg, nextImg, prevPts, nextPts)

View Source

Calculates an optical flow for a sparse feature set using the iterative Lucas-Kanade method with pyramids.

Positional Arguments
  • prevImg: Evision.Mat.t().

    first 8-bit input image or pyramid constructed by buildOpticalFlowPyramid.

  • nextImg: Evision.Mat.t().

    second input image or pyramid of the same size and the same type as prevImg.

  • prevPts: Evision.Mat.t().

    vector of 2D points for which the flow needs to be found; point coordinates must be single-precision floating-point numbers.

Keyword Arguments
  • winSize: Size.

    size of the search window at each pyramid level.

  • maxLevel: int.

    0-based maximal pyramid level number; if set to 0, pyramids are not used (single level), if set to 1, two levels are used, and so on; if pyramids are passed to input then algorithm will use as many levels as pyramids have but no more than maxLevel.

  • criteria: TermCriteria.

    parameter, specifying the termination criteria of the iterative search algorithm (after the specified maximum number of iterations criteria.maxCount or when the search window moves by less than criteria.epsilon.

  • flags: int.

    operation flags:

    • OPTFLOW_USE_INITIAL_FLOW uses initial estimations, stored in nextPts; if the flag is not set, then prevPts is copied to nextPts and is considered the initial estimate.
    • OPTFLOW_LK_GET_MIN_EIGENVALS use minimum eigen values as an error measure (see minEigThreshold description); if the flag is not set, then L1 distance between patches around the original and a moved point, divided by number of pixels in a window, is used as a error measure.
  • minEigThreshold: double.

    the algorithm calculates the minimum eigen value of a 2x2 normal matrix of optical flow equations (this matrix is called a spatial gradient matrix in @cite Bouguet00), divided by number of pixels in a window; if this value is less than minEigThreshold, then a corresponding feature is filtered out and its flow is not processed, so it allows to remove bad points and get a performance boost.

Return
  • nextPts: Evision.Mat.t().

    output vector of 2D points (with single-precision floating-point coordinates) containing the calculated new positions of input features in the second image; when OPTFLOW_USE_INITIAL_FLOW flag is passed, the vector must have the same size as in the input.

  • status: Evision.Mat.t().

    output status vector (of unsigned chars); each element of the vector is set to 1 if the flow for the corresponding features has been found, otherwise, it is set to 0.

  • err: Evision.Mat.t().

    output vector of errors; each element of the vector is set to an error for the corresponding feature, type of the error measure can be set in flags parameter; if the flow wasn't found then the error is not defined (use the status parameter to find such cases).

The function implements a sparse iterative version of the Lucas-Kanade optical flow in pyramids. See @cite Bouguet00 . The function is parallelized with the TBB library. Note: Some examples:

  • An example using the Lucas-Kanade optical flow algorithm can be found at opencv_source_code/samples/cpp/lkdemo.cpp

  • (Python) An example using the Lucas-Kanade optical flow algorithm can be found at opencv_source_code/samples/python/lk_track.py

  • (Python) An example using the Lucas-Kanade tracker for homography matching can be found at opencv_source_code/samples/python/lk_homography.py

Python prototype (for reference only):

calcOpticalFlowPyrLK(prevImg, nextImg, prevPts, nextPts[, status[, err[, winSize[, maxLevel[, criteria[, flags[, minEigThreshold]]]]]]]) -> nextPts, status, err
Link to this function

calcOpticalFlowPyrLK(prevImg, nextImg, prevPts, nextPts, opts)

View Source

Calculates an optical flow for a sparse feature set using the iterative Lucas-Kanade method with pyramids.

Positional Arguments
  • prevImg: Evision.Mat.t().

    first 8-bit input image or pyramid constructed by buildOpticalFlowPyramid.

  • nextImg: Evision.Mat.t().

    second input image or pyramid of the same size and the same type as prevImg.

  • prevPts: Evision.Mat.t().

    vector of 2D points for which the flow needs to be found; point coordinates must be single-precision floating-point numbers.

Keyword Arguments
  • winSize: Size.

    size of the search window at each pyramid level.

  • maxLevel: int.

    0-based maximal pyramid level number; if set to 0, pyramids are not used (single level), if set to 1, two levels are used, and so on; if pyramids are passed to input then algorithm will use as many levels as pyramids have but no more than maxLevel.

  • criteria: TermCriteria.

    parameter, specifying the termination criteria of the iterative search algorithm (after the specified maximum number of iterations criteria.maxCount or when the search window moves by less than criteria.epsilon.

  • flags: int.

    operation flags:

    • OPTFLOW_USE_INITIAL_FLOW uses initial estimations, stored in nextPts; if the flag is not set, then prevPts is copied to nextPts and is considered the initial estimate.
    • OPTFLOW_LK_GET_MIN_EIGENVALS use minimum eigen values as an error measure (see minEigThreshold description); if the flag is not set, then L1 distance between patches around the original and a moved point, divided by number of pixels in a window, is used as a error measure.
  • minEigThreshold: double.

    the algorithm calculates the minimum eigen value of a 2x2 normal matrix of optical flow equations (this matrix is called a spatial gradient matrix in @cite Bouguet00), divided by number of pixels in a window; if this value is less than minEigThreshold, then a corresponding feature is filtered out and its flow is not processed, so it allows to remove bad points and get a performance boost.

Return
  • nextPts: Evision.Mat.t().

    output vector of 2D points (with single-precision floating-point coordinates) containing the calculated new positions of input features in the second image; when OPTFLOW_USE_INITIAL_FLOW flag is passed, the vector must have the same size as in the input.

  • status: Evision.Mat.t().

    output status vector (of unsigned chars); each element of the vector is set to 1 if the flow for the corresponding features has been found, otherwise, it is set to 0.

  • err: Evision.Mat.t().

    output vector of errors; each element of the vector is set to an error for the corresponding feature, type of the error measure can be set in flags parameter; if the flow wasn't found then the error is not defined (use the status parameter to find such cases).

The function implements a sparse iterative version of the Lucas-Kanade optical flow in pyramids. See @cite Bouguet00 . The function is parallelized with the TBB library. Note: Some examples:

  • An example using the Lucas-Kanade optical flow algorithm can be found at opencv_source_code/samples/cpp/lkdemo.cpp

  • (Python) An example using the Lucas-Kanade optical flow algorithm can be found at opencv_source_code/samples/python/lk_track.py

  • (Python) An example using the Lucas-Kanade tracker for homography matching can be found at opencv_source_code/samples/python/lk_homography.py

Python prototype (for reference only):

calcOpticalFlowPyrLK(prevImg, nextImg, prevPts, nextPts[, status[, err[, winSize[, maxLevel[, criteria[, flags[, minEigThreshold]]]]]]]) -> nextPts, status, err
Link to this function

calibrateCamera(objectPoints, imagePoints, imageSize, cameraMatrix, distCoeffs)

View Source

calibrateCamera

Positional Arguments
  • objectPoints: [Evision.Mat]
  • imagePoints: [Evision.Mat]
  • imageSize: Size
Keyword Arguments
  • flags: int.
  • criteria: TermCriteria.
Return
  • retval: double
  • cameraMatrix: Evision.Mat.t()
  • distCoeffs: Evision.Mat.t()
  • rvecs: [Evision.Mat].
  • tvecs: [Evision.Mat].

Has overloading in C++

Python prototype (for reference only):

calibrateCamera(objectPoints, imagePoints, imageSize, cameraMatrix, distCoeffs[, rvecs[, tvecs[, flags[, criteria]]]]) -> retval, cameraMatrix, distCoeffs, rvecs, tvecs
Link to this function

calibrateCamera(objectPoints, imagePoints, imageSize, cameraMatrix, distCoeffs, opts)

View Source

calibrateCamera

Positional Arguments
  • objectPoints: [Evision.Mat]
  • imagePoints: [Evision.Mat]
  • imageSize: Size
Keyword Arguments
  • flags: int.
  • criteria: TermCriteria.
Return
  • retval: double
  • cameraMatrix: Evision.Mat.t()
  • distCoeffs: Evision.Mat.t()
  • rvecs: [Evision.Mat].
  • tvecs: [Evision.Mat].

Has overloading in C++

Python prototype (for reference only):

calibrateCamera(objectPoints, imagePoints, imageSize, cameraMatrix, distCoeffs[, rvecs[, tvecs[, flags[, criteria]]]]) -> retval, cameraMatrix, distCoeffs, rvecs, tvecs
Link to this function

calibrateCameraExtended(objectPoints, imagePoints, imageSize, cameraMatrix, distCoeffs)

View Source

Finds the camera intrinsic and extrinsic parameters from several views of a calibration pattern.

Positional Arguments
  • objectPoints: [Evision.Mat].

    In the new interface it is a vector of vectors of calibration pattern points in the calibration pattern coordinate space (e.g. std::vector<std::vector<cv::Vec3f>>). The outer vector contains as many elements as the number of pattern views. If the same calibration pattern is shown in each view and it is fully visible, all the vectors will be the same. Although, it is possible to use partially occluded patterns or even different patterns in different views. Then, the vectors will be different. Although the points are 3D, they all lie in the calibration pattern's XY coordinate plane (thus 0 in the Z-coordinate), if the used calibration pattern is a planar rig. In the old interface all the vectors of object points from different views are concatenated together.

  • imagePoints: [Evision.Mat].

    In the new interface it is a vector of vectors of the projections of calibration pattern points (e.g. std::vector<std::vector<cv::Vec2f>>). imagePoints.size() and objectPoints.size(), and imagePoints[i].size() and objectPoints[i].size() for each i, must be equal, respectively. In the old interface all the vectors of object points from different views are concatenated together.

  • imageSize: Size.

    Size of the image used only to initialize the camera intrinsic matrix.

Keyword Arguments
  • flags: int.

    Different flags that may be zero or a combination of the following values:

    • @ref CALIB_USE_INTRINSIC_GUESS cameraMatrix contains valid initial values of fx, fy, cx, cy that are optimized further. Otherwise, (cx, cy) is initially set to the image center ( imageSize is used), and focal distances are computed in a least-squares fashion. Note, that if intrinsic parameters are known, there is no need to use this function just to estimate extrinsic parameters. Use @ref solvePnP instead.
    • @ref CALIB_FIX_PRINCIPAL_POINT The principal point is not changed during the global optimization. It stays at the center or at a different location specified when
  • criteria: TermCriteria.

    Termination criteria for the iterative optimization algorithm.

Return
  • retval: double

  • cameraMatrix: Evision.Mat.t().

    Input/output 3x3 floating-point camera intrinsic matrix \f$\cameramatrix{A}\f$ . If @ref CALIB_USE_INTRINSIC_GUESS and/or @ref CALIB_FIX_ASPECT_RATIO, @ref CALIB_FIX_PRINCIPAL_POINT or @ref CALIB_FIX_FOCAL_LENGTH are specified, some or all of fx, fy, cx, cy must be initialized before calling the function.

  • distCoeffs: Evision.Mat.t().

    Input/output vector of distortion coefficients \f$\distcoeffs\f$.

  • rvecs: [Evision.Mat].

    Output vector of rotation vectors (@ref Rodrigues ) estimated for each pattern view (e.g. std::vector<cv::Mat>>). That is, each i-th rotation vector together with the corresponding i-th translation vector (see the next output parameter description) brings the calibration pattern from the object coordinate space (in which object points are specified) to the camera coordinate space. In more technical terms, the tuple of the i-th rotation and translation vector performs a change of basis from object coordinate space to camera coordinate space. Due to its duality, this tuple is equivalent to the position of the calibration pattern with respect to the camera coordinate space.

  • tvecs: [Evision.Mat].

    Output vector of translation vectors estimated for each pattern view, see parameter describtion above.

  • stdDeviationsIntrinsics: Evision.Mat.t().

    Output vector of standard deviations estimated for intrinsic parameters. Order of deviations values: \f$(f_x, f_y, c_x, c_y, k_1, k_2, p_1, p_2, k_3, k_4, k_5, k_6 , s_1, s_2, s_3, s_4, \tau_x, \tau_y)\f$ If one of parameters is not estimated, it's deviation is equals to zero.

  • stdDeviationsExtrinsics: Evision.Mat.t().

    Output vector of standard deviations estimated for extrinsic parameters. Order of deviations values: \f$(R0, T_0, \dotsc , R{M - 1}, T_{M - 1})\f$ where M is the number of pattern views. \f$R_i, T_i\f$ are concatenated 1x3 vectors.

  • perViewErrors: Evision.Mat.t().

    Output vector of the RMS re-projection error estimated for each pattern view.

@ref CALIB_USE_INTRINSIC_GUESS is set too.

  • @ref CALIB_FIX_ASPECT_RATIO The functions consider only fy as a free parameter. The ratio fx/fy stays the same as in the input cameraMatrix . When

@ref CALIB_USE_INTRINSIC_GUESS is not set, the actual input values of fx and fy are ignored, only their ratio is computed and used further.

  • @ref CALIB_ZERO_TANGENT_DIST Tangential distortion coefficients \f$(p_1, p_2)\f$ are set to zeros and stay zero.

  • @ref CALIB_FIX_FOCAL_LENGTH The focal length is not changed during the global optimization if @ref CALIB_USE_INTRINSIC_GUESS is set.

  • @ref CALIB_FIX_K1,..., @ref CALIB_FIX_K6 The corresponding radial distortion coefficient is not changed during the optimization. If @ref CALIB_USE_INTRINSIC_GUESS is set, the coefficient from the supplied distCoeffs matrix is used. Otherwise, it is set to 0.

  • @ref CALIB_RATIONAL_MODEL Coefficients k4, k5, and k6 are enabled. To provide the backward compatibility, this extra flag should be explicitly specified to make the calibration function use the rational model and return 8 coefficients or more.

  • @ref CALIB_THIN_PRISM_MODEL Coefficients s1, s2, s3 and s4 are enabled. To provide the backward compatibility, this extra flag should be explicitly specified to make the calibration function use the thin prism model and return 12 coefficients or more.

  • @ref CALIB_FIX_S1_S2_S3_S4 The thin prism distortion coefficients are not changed during the optimization. If @ref CALIB_USE_INTRINSIC_GUESS is set, the coefficient from the supplied distCoeffs matrix is used. Otherwise, it is set to 0.

  • @ref CALIB_TILTED_MODEL Coefficients tauX and tauY are enabled. To provide the backward compatibility, this extra flag should be explicitly specified to make the calibration function use the tilted sensor model and return 14 coefficients.

  • @ref CALIB_FIX_TAUX_TAUY The coefficients of the tilted sensor model are not changed during the optimization. If @ref CALIB_USE_INTRINSIC_GUESS is set, the coefficient from the supplied distCoeffs matrix is used. Otherwise, it is set to 0.

@return the overall RMS re-projection error. The function estimates the intrinsic camera parameters and extrinsic parameters for each of the views. The algorithm is based on @cite Zhang2000 and @cite BouguetMCT . The coordinates of 3D object points and their corresponding 2D projections in each view must be specified. That may be achieved by using an object with known geometry and easily detectable feature points. Such an object is called a calibration rig or calibration pattern, and OpenCV has built-in support for a chessboard as a calibration rig (see @ref findChessboardCorners). Currently, initialization of intrinsic parameters (when @ref CALIB_USE_INTRINSIC_GUESS is not set) is only implemented for planar calibration patterns (where Z-coordinates of the object points must be all zeros). 3D calibration rigs can also be used as long as initial cameraMatrix is provided. The algorithm performs the following steps:

  • Compute the initial intrinsic parameters (the option only available for planar calibration patterns) or read them from the input parameters. The distortion coefficients are all set to zeros initially unless some of CALIB_FIX_K? are specified.

  • Estimate the initial camera pose as if the intrinsic parameters have been already known. This is done using @ref solvePnP .

  • Run the global Levenberg-Marquardt optimization algorithm to minimize the reprojection error, that is, the total sum of squared distances between the observed feature points imagePoints and the projected (using the current estimates for camera parameters and the poses) object points objectPoints. See @ref projectPoints for details.

Note: If you use a non-square (i.e. non-N-by-N) grid and @ref findChessboardCorners for calibration, and @ref calibrateCamera returns bad values (zero distortion coefficients, \f$c_x\f$ and \f$c_y\f$ very far from the image center, and/or large differences between \f$f_x\f$ and \f$f_y\f$ (ratios of 10:1 or more)), then you are probably using patternSize=cvSize(rows,cols) instead of using patternSize=cvSize(cols,rows) in @ref findChessboardCorners. Note: The function may throw exceptions, if unsupported combination of parameters is provided or the system is underconstrained. @sa calibrateCameraRO, findChessboardCorners, solvePnP, initCameraMatrix2D, stereoCalibrate, undistort

Python prototype (for reference only):

calibrateCameraExtended(objectPoints, imagePoints, imageSize, cameraMatrix, distCoeffs[, rvecs[, tvecs[, stdDeviationsIntrinsics[, stdDeviationsExtrinsics[, perViewErrors[, flags[, criteria]]]]]]]) -> retval, cameraMatrix, distCoeffs, rvecs, tvecs, stdDeviationsIntrinsics, stdDeviationsExtrinsics, perViewErrors
Link to this function

calibrateCameraExtended(objectPoints, imagePoints, imageSize, cameraMatrix, distCoeffs, opts)

View Source

Finds the camera intrinsic and extrinsic parameters from several views of a calibration pattern.

Positional Arguments
  • objectPoints: [Evision.Mat].

    In the new interface it is a vector of vectors of calibration pattern points in the calibration pattern coordinate space (e.g. std::vector<std::vector<cv::Vec3f>>). The outer vector contains as many elements as the number of pattern views. If the same calibration pattern is shown in each view and it is fully visible, all the vectors will be the same. Although, it is possible to use partially occluded patterns or even different patterns in different views. Then, the vectors will be different. Although the points are 3D, they all lie in the calibration pattern's XY coordinate plane (thus 0 in the Z-coordinate), if the used calibration pattern is a planar rig. In the old interface all the vectors of object points from different views are concatenated together.

  • imagePoints: [Evision.Mat].

    In the new interface it is a vector of vectors of the projections of calibration pattern points (e.g. std::vector<std::vector<cv::Vec2f>>). imagePoints.size() and objectPoints.size(), and imagePoints[i].size() and objectPoints[i].size() for each i, must be equal, respectively. In the old interface all the vectors of object points from different views are concatenated together.

  • imageSize: Size.

    Size of the image used only to initialize the camera intrinsic matrix.

Keyword Arguments
  • flags: int.

    Different flags that may be zero or a combination of the following values:

    • @ref CALIB_USE_INTRINSIC_GUESS cameraMatrix contains valid initial values of fx, fy, cx, cy that are optimized further. Otherwise, (cx, cy) is initially set to the image center ( imageSize is used), and focal distances are computed in a least-squares fashion. Note, that if intrinsic parameters are known, there is no need to use this function just to estimate extrinsic parameters. Use @ref solvePnP instead.
    • @ref CALIB_FIX_PRINCIPAL_POINT The principal point is not changed during the global optimization. It stays at the center or at a different location specified when
  • criteria: TermCriteria.

    Termination criteria for the iterative optimization algorithm.

Return
  • retval: double

  • cameraMatrix: Evision.Mat.t().

    Input/output 3x3 floating-point camera intrinsic matrix \f$\cameramatrix{A}\f$ . If @ref CALIB_USE_INTRINSIC_GUESS and/or @ref CALIB_FIX_ASPECT_RATIO, @ref CALIB_FIX_PRINCIPAL_POINT or @ref CALIB_FIX_FOCAL_LENGTH are specified, some or all of fx, fy, cx, cy must be initialized before calling the function.

  • distCoeffs: Evision.Mat.t().

    Input/output vector of distortion coefficients \f$\distcoeffs\f$.

  • rvecs: [Evision.Mat].

    Output vector of rotation vectors (@ref Rodrigues ) estimated for each pattern view (e.g. std::vector<cv::Mat>>). That is, each i-th rotation vector together with the corresponding i-th translation vector (see the next output parameter description) brings the calibration pattern from the object coordinate space (in which object points are specified) to the camera coordinate space. In more technical terms, the tuple of the i-th rotation and translation vector performs a change of basis from object coordinate space to camera coordinate space. Due to its duality, this tuple is equivalent to the position of the calibration pattern with respect to the camera coordinate space.

  • tvecs: [Evision.Mat].

    Output vector of translation vectors estimated for each pattern view, see parameter describtion above.

  • stdDeviationsIntrinsics: Evision.Mat.t().

    Output vector of standard deviations estimated for intrinsic parameters. Order of deviations values: \f$(f_x, f_y, c_x, c_y, k_1, k_2, p_1, p_2, k_3, k_4, k_5, k_6 , s_1, s_2, s_3, s_4, \tau_x, \tau_y)\f$ If one of parameters is not estimated, it's deviation is equals to zero.

  • stdDeviationsExtrinsics: Evision.Mat.t().

    Output vector of standard deviations estimated for extrinsic parameters. Order of deviations values: \f$(R0, T_0, \dotsc , R{M - 1}, T_{M - 1})\f$ where M is the number of pattern views. \f$R_i, T_i\f$ are concatenated 1x3 vectors.

  • perViewErrors: Evision.Mat.t().

    Output vector of the RMS re-projection error estimated for each pattern view.

@ref CALIB_USE_INTRINSIC_GUESS is set too.

  • @ref CALIB_FIX_ASPECT_RATIO The functions consider only fy as a free parameter. The ratio fx/fy stays the same as in the input cameraMatrix . When

@ref CALIB_USE_INTRINSIC_GUESS is not set, the actual input values of fx and fy are ignored, only their ratio is computed and used further.

  • @ref CALIB_ZERO_TANGENT_DIST Tangential distortion coefficients \f$(p_1, p_2)\f$ are set to zeros and stay zero.

  • @ref CALIB_FIX_FOCAL_LENGTH The focal length is not changed during the global optimization if @ref CALIB_USE_INTRINSIC_GUESS is set.

  • @ref CALIB_FIX_K1,..., @ref CALIB_FIX_K6 The corresponding radial distortion coefficient is not changed during the optimization. If @ref CALIB_USE_INTRINSIC_GUESS is set, the coefficient from the supplied distCoeffs matrix is used. Otherwise, it is set to 0.

  • @ref CALIB_RATIONAL_MODEL Coefficients k4, k5, and k6 are enabled. To provide the backward compatibility, this extra flag should be explicitly specified to make the calibration function use the rational model and return 8 coefficients or more.

  • @ref CALIB_THIN_PRISM_MODEL Coefficients s1, s2, s3 and s4 are enabled. To provide the backward compatibility, this extra flag should be explicitly specified to make the calibration function use the thin prism model and return 12 coefficients or more.

  • @ref CALIB_FIX_S1_S2_S3_S4 The thin prism distortion coefficients are not changed during the optimization. If @ref CALIB_USE_INTRINSIC_GUESS is set, the coefficient from the supplied distCoeffs matrix is used. Otherwise, it is set to 0.

  • @ref CALIB_TILTED_MODEL Coefficients tauX and tauY are enabled. To provide the backward compatibility, this extra flag should be explicitly specified to make the calibration function use the tilted sensor model and return 14 coefficients.

  • @ref CALIB_FIX_TAUX_TAUY The coefficients of the tilted sensor model are not changed during the optimization. If @ref CALIB_USE_INTRINSIC_GUESS is set, the coefficient from the supplied distCoeffs matrix is used. Otherwise, it is set to 0.

@return the overall RMS re-projection error. The function estimates the intrinsic camera parameters and extrinsic parameters for each of the views. The algorithm is based on @cite Zhang2000 and @cite BouguetMCT . The coordinates of 3D object points and their corresponding 2D projections in each view must be specified. That may be achieved by using an object with known geometry and easily detectable feature points. Such an object is called a calibration rig or calibration pattern, and OpenCV has built-in support for a chessboard as a calibration rig (see @ref findChessboardCorners). Currently, initialization of intrinsic parameters (when @ref CALIB_USE_INTRINSIC_GUESS is not set) is only implemented for planar calibration patterns (where Z-coordinates of the object points must be all zeros). 3D calibration rigs can also be used as long as initial cameraMatrix is provided. The algorithm performs the following steps:

  • Compute the initial intrinsic parameters (the option only available for planar calibration patterns) or read them from the input parameters. The distortion coefficients are all set to zeros initially unless some of CALIB_FIX_K? are specified.

  • Estimate the initial camera pose as if the intrinsic parameters have been already known. This is done using @ref solvePnP .

  • Run the global Levenberg-Marquardt optimization algorithm to minimize the reprojection error, that is, the total sum of squared distances between the observed feature points imagePoints and the projected (using the current estimates for camera parameters and the poses) object points objectPoints. See @ref projectPoints for details.

Note: If you use a non-square (i.e. non-N-by-N) grid and @ref findChessboardCorners for calibration, and @ref calibrateCamera returns bad values (zero distortion coefficients, \f$c_x\f$ and \f$c_y\f$ very far from the image center, and/or large differences between \f$f_x\f$ and \f$f_y\f$ (ratios of 10:1 or more)), then you are probably using patternSize=cvSize(rows,cols) instead of using patternSize=cvSize(cols,rows) in @ref findChessboardCorners. Note: The function may throw exceptions, if unsupported combination of parameters is provided or the system is underconstrained. @sa calibrateCameraRO, findChessboardCorners, solvePnP, initCameraMatrix2D, stereoCalibrate, undistort

Python prototype (for reference only):

calibrateCameraExtended(objectPoints, imagePoints, imageSize, cameraMatrix, distCoeffs[, rvecs[, tvecs[, stdDeviationsIntrinsics[, stdDeviationsExtrinsics[, perViewErrors[, flags[, criteria]]]]]]]) -> retval, cameraMatrix, distCoeffs, rvecs, tvecs, stdDeviationsIntrinsics, stdDeviationsExtrinsics, perViewErrors
Link to this function

calibrateCameraRO(objectPoints, imagePoints, imageSize, iFixedPoint, cameraMatrix, distCoeffs)

View Source

calibrateCameraRO

Positional Arguments
  • objectPoints: [Evision.Mat]
  • imagePoints: [Evision.Mat]
  • imageSize: Size
  • iFixedPoint: int
Keyword Arguments
  • flags: int.
  • criteria: TermCriteria.
Return
  • retval: double
  • cameraMatrix: Evision.Mat.t()
  • distCoeffs: Evision.Mat.t()
  • rvecs: [Evision.Mat].
  • tvecs: [Evision.Mat].
  • newObjPoints: Evision.Mat.t().

Has overloading in C++

Python prototype (for reference only):

calibrateCameraRO(objectPoints, imagePoints, imageSize, iFixedPoint, cameraMatrix, distCoeffs[, rvecs[, tvecs[, newObjPoints[, flags[, criteria]]]]]) -> retval, cameraMatrix, distCoeffs, rvecs, tvecs, newObjPoints
Link to this function

calibrateCameraRO(objectPoints, imagePoints, imageSize, iFixedPoint, cameraMatrix, distCoeffs, opts)

View Source

calibrateCameraRO

Positional Arguments
  • objectPoints: [Evision.Mat]
  • imagePoints: [Evision.Mat]
  • imageSize: Size
  • iFixedPoint: int
Keyword Arguments
  • flags: int.
  • criteria: TermCriteria.
Return
  • retval: double
  • cameraMatrix: Evision.Mat.t()
  • distCoeffs: Evision.Mat.t()
  • rvecs: [Evision.Mat].
  • tvecs: [Evision.Mat].
  • newObjPoints: Evision.Mat.t().

Has overloading in C++

Python prototype (for reference only):

calibrateCameraRO(objectPoints, imagePoints, imageSize, iFixedPoint, cameraMatrix, distCoeffs[, rvecs[, tvecs[, newObjPoints[, flags[, criteria]]]]]) -> retval, cameraMatrix, distCoeffs, rvecs, tvecs, newObjPoints
Link to this function

calibrateCameraROExtended(objectPoints, imagePoints, imageSize, iFixedPoint, cameraMatrix, distCoeffs)

View Source

Finds the camera intrinsic and extrinsic parameters from several views of a calibration pattern.

Positional Arguments
  • objectPoints: [Evision.Mat].

    Vector of vectors of calibration pattern points in the calibration pattern coordinate space. See #calibrateCamera for details. If the method of releasing object to be used, the identical calibration board must be used in each view and it must be fully visible, and all objectPoints[i] must be the same and all points should be roughly close to a plane. The calibration target has to be rigid, or at least static if the camera (rather than the calibration target) is shifted for grabbing images.

  • imagePoints: [Evision.Mat].

    Vector of vectors of the projections of calibration pattern points. See #calibrateCamera for details.

  • imageSize: Size.

    Size of the image used only to initialize the intrinsic camera matrix.

  • iFixedPoint: int.

    The index of the 3D object point in objectPoints[0] to be fixed. It also acts as a switch for calibration method selection. If object-releasing method to be used, pass in the parameter in the range of [1, objectPoints[0].size()-2], otherwise a value out of this range will make standard calibration method selected. Usually the top-right corner point of the calibration board grid is recommended to be fixed when object-releasing method being utilized. According to \cite strobl2011iccv, two other points are also fixed. In this implementation, objectPoints[0].front and objectPoints[0].back.z are used. With object-releasing method, accurate rvecs, tvecs and newObjPoints are only possible if coordinates of these three fixed points are accurate enough.

Keyword Arguments
  • flags: int.

    Different flags that may be zero or a combination of some predefined values. See #calibrateCamera for details. If the method of releasing object is used, the calibration time may be much longer. CALIB_USE_QR or CALIB_USE_LU could be used for faster calibration with potentially less precise and less stable in some rare cases.

  • criteria: TermCriteria.

    Termination criteria for the iterative optimization algorithm.

Return
  • retval: double

  • cameraMatrix: Evision.Mat.t().

    Output 3x3 floating-point camera matrix. See #calibrateCamera for details.

  • distCoeffs: Evision.Mat.t().

    Output vector of distortion coefficients. See #calibrateCamera for details.

  • rvecs: [Evision.Mat].

    Output vector of rotation vectors estimated for each pattern view. See #calibrateCamera for details.

  • tvecs: [Evision.Mat].

    Output vector of translation vectors estimated for each pattern view.

  • newObjPoints: Evision.Mat.t().

    The updated output vector of calibration pattern points. The coordinates might be scaled based on three fixed points. The returned coordinates are accurate only if the above mentioned three fixed points are accurate. If not needed, noArray() can be passed in. This parameter is ignored with standard calibration method.

  • stdDeviationsIntrinsics: Evision.Mat.t().

    Output vector of standard deviations estimated for intrinsic parameters. See #calibrateCamera for details.

  • stdDeviationsExtrinsics: Evision.Mat.t().

    Output vector of standard deviations estimated for extrinsic parameters. See #calibrateCamera for details.

  • stdDeviationsObjPoints: Evision.Mat.t().

    Output vector of standard deviations estimated for refined coordinates of calibration pattern points. It has the same size and order as objectPoints[0] vector. This parameter is ignored with standard calibration method.

  • perViewErrors: Evision.Mat.t().

    Output vector of the RMS re-projection error estimated for each pattern view.

This function is an extension of #calibrateCamera with the method of releasing object which was proposed in @cite strobl2011iccv. In many common cases with inaccurate, unmeasured, roughly planar targets (calibration plates), this method can dramatically improve the precision of the estimated camera parameters. Both the object-releasing method and standard method are supported by this function. Use the parameter iFixedPoint for method selection. In the internal implementation, #calibrateCamera is a wrapper for this function.

@return the overall RMS re-projection error. The function estimates the intrinsic camera parameters and extrinsic parameters for each of the views. The algorithm is based on @cite Zhang2000, @cite BouguetMCT and @cite strobl2011iccv. See #calibrateCamera for other detailed explanations. @sa calibrateCamera, findChessboardCorners, solvePnP, initCameraMatrix2D, stereoCalibrate, undistort

Python prototype (for reference only):

calibrateCameraROExtended(objectPoints, imagePoints, imageSize, iFixedPoint, cameraMatrix, distCoeffs[, rvecs[, tvecs[, newObjPoints[, stdDeviationsIntrinsics[, stdDeviationsExtrinsics[, stdDeviationsObjPoints[, perViewErrors[, flags[, criteria]]]]]]]]]) -> retval, cameraMatrix, distCoeffs, rvecs, tvecs, newObjPoints, stdDeviationsIntrinsics, stdDeviationsExtrinsics, stdDeviationsObjPoints, perViewErrors
Link to this function

calibrateCameraROExtended(objectPoints, imagePoints, imageSize, iFixedPoint, cameraMatrix, distCoeffs, opts)

View Source

Finds the camera intrinsic and extrinsic parameters from several views of a calibration pattern.

Positional Arguments
  • objectPoints: [Evision.Mat].

    Vector of vectors of calibration pattern points in the calibration pattern coordinate space. See #calibrateCamera for details. If the method of releasing object to be used, the identical calibration board must be used in each view and it must be fully visible, and all objectPoints[i] must be the same and all points should be roughly close to a plane. The calibration target has to be rigid, or at least static if the camera (rather than the calibration target) is shifted for grabbing images.

  • imagePoints: [Evision.Mat].

    Vector of vectors of the projections of calibration pattern points. See #calibrateCamera for details.

  • imageSize: Size.

    Size of the image used only to initialize the intrinsic camera matrix.

  • iFixedPoint: int.

    The index of the 3D object point in objectPoints[0] to be fixed. It also acts as a switch for calibration method selection. If object-releasing method to be used, pass in the parameter in the range of [1, objectPoints[0].size()-2], otherwise a value out of this range will make standard calibration method selected. Usually the top-right corner point of the calibration board grid is recommended to be fixed when object-releasing method being utilized. According to \cite strobl2011iccv, two other points are also fixed. In this implementation, objectPoints[0].front and objectPoints[0].back.z are used. With object-releasing method, accurate rvecs, tvecs and newObjPoints are only possible if coordinates of these three fixed points are accurate enough.

Keyword Arguments
  • flags: int.

    Different flags that may be zero or a combination of some predefined values. See #calibrateCamera for details. If the method of releasing object is used, the calibration time may be much longer. CALIB_USE_QR or CALIB_USE_LU could be used for faster calibration with potentially less precise and less stable in some rare cases.

  • criteria: TermCriteria.

    Termination criteria for the iterative optimization algorithm.

Return
  • retval: double

  • cameraMatrix: Evision.Mat.t().

    Output 3x3 floating-point camera matrix. See #calibrateCamera for details.

  • distCoeffs: Evision.Mat.t().

    Output vector of distortion coefficients. See #calibrateCamera for details.

  • rvecs: [Evision.Mat].

    Output vector of rotation vectors estimated for each pattern view. See #calibrateCamera for details.

  • tvecs: [Evision.Mat].

    Output vector of translation vectors estimated for each pattern view.

  • newObjPoints: Evision.Mat.t().

    The updated output vector of calibration pattern points. The coordinates might be scaled based on three fixed points. The returned coordinates are accurate only if the above mentioned three fixed points are accurate. If not needed, noArray() can be passed in. This parameter is ignored with standard calibration method.

  • stdDeviationsIntrinsics: Evision.Mat.t().

    Output vector of standard deviations estimated for intrinsic parameters. See #calibrateCamera for details.

  • stdDeviationsExtrinsics: Evision.Mat.t().

    Output vector of standard deviations estimated for extrinsic parameters. See #calibrateCamera for details.

  • stdDeviationsObjPoints: Evision.Mat.t().

    Output vector of standard deviations estimated for refined coordinates of calibration pattern points. It has the same size and order as objectPoints[0] vector. This parameter is ignored with standard calibration method.

  • perViewErrors: Evision.Mat.t().

    Output vector of the RMS re-projection error estimated for each pattern view.

This function is an extension of #calibrateCamera with the method of releasing object which was proposed in @cite strobl2011iccv. In many common cases with inaccurate, unmeasured, roughly planar targets (calibration plates), this method can dramatically improve the precision of the estimated camera parameters. Both the object-releasing method and standard method are supported by this function. Use the parameter iFixedPoint for method selection. In the internal implementation, #calibrateCamera is a wrapper for this function.

@return the overall RMS re-projection error. The function estimates the intrinsic camera parameters and extrinsic parameters for each of the views. The algorithm is based on @cite Zhang2000, @cite BouguetMCT and @cite strobl2011iccv. See #calibrateCamera for other detailed explanations. @sa calibrateCamera, findChessboardCorners, solvePnP, initCameraMatrix2D, stereoCalibrate, undistort

Python prototype (for reference only):

calibrateCameraROExtended(objectPoints, imagePoints, imageSize, iFixedPoint, cameraMatrix, distCoeffs[, rvecs[, tvecs[, newObjPoints[, stdDeviationsIntrinsics[, stdDeviationsExtrinsics[, stdDeviationsObjPoints[, perViewErrors[, flags[, criteria]]]]]]]]]) -> retval, cameraMatrix, distCoeffs, rvecs, tvecs, newObjPoints, stdDeviationsIntrinsics, stdDeviationsExtrinsics, stdDeviationsObjPoints, perViewErrors
Link to this function

calibrateHandEye(r_gripper2base, t_gripper2base, r_target2cam, t_target2cam)

View Source

Computes Hand-Eye calibration: \f$_{}^{g}\textrm{T}_c\f$

Positional Arguments
  • r_gripper2base: [Evision.Mat].

    Rotation part extracted from the homogeneous matrix that transforms a point expressed in the gripper frame to the robot base frame (\f$_{}^{b}\textrm{T}_g\f$). This is a vector (vector<Mat>) that contains the rotation, (3x3) rotation matrices or (3x1) rotation vectors, for all the transformations from gripper frame to robot base frame.

  • t_gripper2base: [Evision.Mat].

    Translation part extracted from the homogeneous matrix that transforms a point expressed in the gripper frame to the robot base frame (\f$_{}^{b}\textrm{T}_g\f$). This is a vector (vector<Mat>) that contains the (3x1) translation vectors for all the transformations from gripper frame to robot base frame.

  • r_target2cam: [Evision.Mat].

    Rotation part extracted from the homogeneous matrix that transforms a point expressed in the target frame to the camera frame (\f$_{}^{c}\textrm{T}_t\f$). This is a vector (vector<Mat>) that contains the rotation, (3x3) rotation matrices or (3x1) rotation vectors, for all the transformations from calibration target frame to camera frame.

  • t_target2cam: [Evision.Mat].

    Rotation part extracted from the homogeneous matrix that transforms a point expressed in the target frame to the camera frame (\f$_{}^{c}\textrm{T}_t\f$). This is a vector (vector<Mat>) that contains the (3x1) translation vectors for all the transformations from calibration target frame to camera frame.

Keyword Arguments
  • method: HandEyeCalibrationMethod.

    One of the implemented Hand-Eye calibration method, see cv::HandEyeCalibrationMethod

Return
  • r_cam2gripper: Evision.Mat.t().

    Estimated (3x3) rotation part extracted from the homogeneous matrix that transforms a point expressed in the camera frame to the gripper frame (\f$_{}^{g}\textrm{T}_c\f$).

  • t_cam2gripper: Evision.Mat.t().

    Estimated (3x1) translation part extracted from the homogeneous matrix that transforms a point expressed in the camera frame to the gripper frame (\f$_{}^{g}\textrm{T}_c\f$).

The function performs the Hand-Eye calibration using various methods. One approach consists in estimating the rotation then the translation (separable solutions) and the following methods are implemented:

  • R. Tsai, R. Lenz A New Technique for Fully Autonomous and Efficient 3D Robotics Hand/EyeCalibration \cite Tsai89
  • F. Park, B. Martin Robot Sensor Calibration: Solving AX = XB on the Euclidean Group \cite Park94
  • R. Horaud, F. Dornaika Hand-Eye Calibration \cite Horaud95

Another approach consists in estimating simultaneously the rotation and the translation (simultaneous solutions), with the following implemented methods:

  • N. Andreff, R. Horaud, B. Espiau On-line Hand-Eye Calibration \cite Andreff99
  • K. Daniilidis Hand-Eye Calibration Using Dual Quaternions \cite Daniilidis98

The following picture describes the Hand-Eye calibration problem where the transformation between a camera ("eye") mounted on a robot gripper ("hand") has to be estimated. This configuration is called eye-in-hand. The eye-to-hand configuration consists in a static camera observing a calibration pattern mounted on the robot end-effector. The transformation from the camera to the robot base frame can then be estimated by inputting the suitable transformations to the function, see below. The calibration procedure is the following:

  • a static calibration pattern is used to estimate the transformation between the target frame and the camera frame

  • the robot gripper is moved in order to acquire several poses

  • for each pose, the homogeneous transformation between the gripper frame and the robot base frame is recorded using for instance the robot kinematics \f[ \begin{bmatrix} X_b\\ Y_b\\ Z_b\\ 1 \end{bmatrix}= \begin{bmatrix} _{}^{b}\textrm{R}_g & _{}^{b}\textrm{t}_g \\ 0_{1 \times 3} & 1 \end{bmatrix} \begin{bmatrix} X_g\\ Y_g\\ Z_g\\ 1 \end{bmatrix} \f]

  • for each pose, the homogeneous transformation between the calibration target frame and the camera frame is recorded using for instance a pose estimation method (PnP) from 2D-3D point correspondences \f[ \begin{bmatrix} X_c\\ Y_c\\ Z_c\\ 1 \end{bmatrix}= \begin{bmatrix} _{}^{c}\textrm{R}_t & _{}^{c}\textrm{t}_t \\ 0_{1 \times 3} & 1 \end{bmatrix} \begin{bmatrix} X_t\\ Y_t\\ Z_t\\ 1 \end{bmatrix} \f]

The Hand-Eye calibration procedure returns the following homogeneous transformation \f[ \begin{bmatrix} X_g\\ Y_g\\ Z_g\\ 1 \end{bmatrix} \begin{bmatrix} _{}^{g}\textrm{R}_c & _{}^{g}\textrm{t}_c \\ 0_{1 \times 3} & 1 \end{bmatrix} \begin{bmatrix} X_c\\ Y_c\\ Z_c\\ 1 \end{bmatrix} \f] This problem is also known as solving the \f$\mathbf{A}\mathbf{X}=\mathbf{X}\mathbf{B}\f$ equation:

  • for an eye-in-hand configuration \f[ \begin{align*} ^{b}{\textrm{T}_g}^{(1)} \hspace{0.2em} ^{g}\textrm{T}_c \hspace{0.2em} ^{c}{\textrm{T}_t}^{(1)} &= \hspace{0.1em} ^{b}{\textrm{T}_g}^{(2)} \hspace{0.2em} ^{g}\textrm{T}_c \hspace{0.2em} ^{c}{\textrm{T}_t}^{(2)} \\ (^{b}{\textrm{T}_g}^{(2)})^{-1} \hspace{0.2em} ^{b}{\textrm{T}_g}^{(1)} \hspace{0.2em} ^{g}\textrm{T}_c &= \hspace{0.1em} ^{g}\textrm{T}_c \hspace{0.2em} ^{c}{\textrm{T}_t}^{(2)} (^{c}{\textrm{T}_t}^{(1)})^{-1} \\ \textrm{A}_i \textrm{X} &= \textrm{X} \textrm{B}_i \\ \end{align*} \f]

  • for an eye-to-hand configuration \f[ \begin{align*} ^{g}{\textrm{T}_b}^{(1)} \hspace{0.2em} ^{b}\textrm{T}_c \hspace{0.2em} ^{c}{\textrm{T}_t}^{(1)} &= \hspace{0.1em} ^{g}{\textrm{T}_b}^{(2)} \hspace{0.2em} ^{b}\textrm{T}_c \hspace{0.2em} ^{c}{\textrm{T}_t}^{(2)} \\ (^{g}{\textrm{T}_b}^{(2)})^{-1} \hspace{0.2em} ^{g}{\textrm{T}_b}^{(1)} \hspace{0.2em} ^{b}\textrm{T}_c &= \hspace{0.1em} ^{b}\textrm{T}_c \hspace{0.2em} ^{c}{\textrm{T}_t}^{(2)} (^{c}{\textrm{T}_t}^{(1)})^{-1} \\ \textrm{A}_i \textrm{X} &= \textrm{X} \textrm{B}_i \\ \end{align*} \f]

\note Additional information can be found on this website. \note A minimum of 2 motions with non parallel rotation axes are necessary to determine the hand-eye transformation. So at least 3 different poses are required, but it is strongly recommended to use many more poses.

Python prototype (for reference only):

calibrateHandEye(R_gripper2base, t_gripper2base, R_target2cam, t_target2cam[, R_cam2gripper[, t_cam2gripper[, method]]]) -> R_cam2gripper, t_cam2gripper
Link to this function

calibrateHandEye(r_gripper2base, t_gripper2base, r_target2cam, t_target2cam, opts)

View Source
@spec calibrateHandEye(
  [Evision.Mat.maybe_mat_in()],
  [Evision.Mat.maybe_mat_in()],
  [Evision.Mat.maybe_mat_in()],
  [Evision.Mat.maybe_mat_in()],
  [{atom(), term()}, ...] | nil
) :: {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Computes Hand-Eye calibration: \f$_{}^{g}\textrm{T}_c\f$

Positional Arguments
  • r_gripper2base: [Evision.Mat].

    Rotation part extracted from the homogeneous matrix that transforms a point expressed in the gripper frame to the robot base frame (\f$_{}^{b}\textrm{T}_g\f$). This is a vector (vector<Mat>) that contains the rotation, (3x3) rotation matrices or (3x1) rotation vectors, for all the transformations from gripper frame to robot base frame.

  • t_gripper2base: [Evision.Mat].

    Translation part extracted from the homogeneous matrix that transforms a point expressed in the gripper frame to the robot base frame (\f$_{}^{b}\textrm{T}_g\f$). This is a vector (vector<Mat>) that contains the (3x1) translation vectors for all the transformations from gripper frame to robot base frame.

  • r_target2cam: [Evision.Mat].

    Rotation part extracted from the homogeneous matrix that transforms a point expressed in the target frame to the camera frame (\f$_{}^{c}\textrm{T}_t\f$). This is a vector (vector<Mat>) that contains the rotation, (3x3) rotation matrices or (3x1) rotation vectors, for all the transformations from calibration target frame to camera frame.

  • t_target2cam: [Evision.Mat].

    Rotation part extracted from the homogeneous matrix that transforms a point expressed in the target frame to the camera frame (\f$_{}^{c}\textrm{T}_t\f$). This is a vector (vector<Mat>) that contains the (3x1) translation vectors for all the transformations from calibration target frame to camera frame.

Keyword Arguments
  • method: HandEyeCalibrationMethod.

    One of the implemented Hand-Eye calibration method, see cv::HandEyeCalibrationMethod

Return
  • r_cam2gripper: Evision.Mat.t().

    Estimated (3x3) rotation part extracted from the homogeneous matrix that transforms a point expressed in the camera frame to the gripper frame (\f$_{}^{g}\textrm{T}_c\f$).

  • t_cam2gripper: Evision.Mat.t().

    Estimated (3x1) translation part extracted from the homogeneous matrix that transforms a point expressed in the camera frame to the gripper frame (\f$_{}^{g}\textrm{T}_c\f$).

The function performs the Hand-Eye calibration using various methods. One approach consists in estimating the rotation then the translation (separable solutions) and the following methods are implemented:

  • R. Tsai, R. Lenz A New Technique for Fully Autonomous and Efficient 3D Robotics Hand/EyeCalibration \cite Tsai89
  • F. Park, B. Martin Robot Sensor Calibration: Solving AX = XB on the Euclidean Group \cite Park94
  • R. Horaud, F. Dornaika Hand-Eye Calibration \cite Horaud95

Another approach consists in estimating simultaneously the rotation and the translation (simultaneous solutions), with the following implemented methods:

  • N. Andreff, R. Horaud, B. Espiau On-line Hand-Eye Calibration \cite Andreff99
  • K. Daniilidis Hand-Eye Calibration Using Dual Quaternions \cite Daniilidis98

The following picture describes the Hand-Eye calibration problem where the transformation between a camera ("eye") mounted on a robot gripper ("hand") has to be estimated. This configuration is called eye-in-hand. The eye-to-hand configuration consists in a static camera observing a calibration pattern mounted on the robot end-effector. The transformation from the camera to the robot base frame can then be estimated by inputting the suitable transformations to the function, see below. The calibration procedure is the following:

  • a static calibration pattern is used to estimate the transformation between the target frame and the camera frame

  • the robot gripper is moved in order to acquire several poses

  • for each pose, the homogeneous transformation between the gripper frame and the robot base frame is recorded using for instance the robot kinematics \f[ \begin{bmatrix} X_b\\ Y_b\\ Z_b\\ 1 \end{bmatrix}= \begin{bmatrix} _{}^{b}\textrm{R}_g & _{}^{b}\textrm{t}_g \\ 0_{1 \times 3} & 1 \end{bmatrix} \begin{bmatrix} X_g\\ Y_g\\ Z_g\\ 1 \end{bmatrix} \f]

  • for each pose, the homogeneous transformation between the calibration target frame and the camera frame is recorded using for instance a pose estimation method (PnP) from 2D-3D point correspondences \f[ \begin{bmatrix} X_c\\ Y_c\\ Z_c\\ 1 \end{bmatrix}= \begin{bmatrix} _{}^{c}\textrm{R}_t & _{}^{c}\textrm{t}_t \\ 0_{1 \times 3} & 1 \end{bmatrix} \begin{bmatrix} X_t\\ Y_t\\ Z_t\\ 1 \end{bmatrix} \f]

The Hand-Eye calibration procedure returns the following homogeneous transformation \f[ \begin{bmatrix} X_g\\ Y_g\\ Z_g\\ 1 \end{bmatrix} \begin{bmatrix} _{}^{g}\textrm{R}_c & _{}^{g}\textrm{t}_c \\ 0_{1 \times 3} & 1 \end{bmatrix} \begin{bmatrix} X_c\\ Y_c\\ Z_c\\ 1 \end{bmatrix} \f] This problem is also known as solving the \f$\mathbf{A}\mathbf{X}=\mathbf{X}\mathbf{B}\f$ equation:

  • for an eye-in-hand configuration \f[ \begin{align*} ^{b}{\textrm{T}_g}^{(1)} \hspace{0.2em} ^{g}\textrm{T}_c \hspace{0.2em} ^{c}{\textrm{T}_t}^{(1)} &= \hspace{0.1em} ^{b}{\textrm{T}_g}^{(2)} \hspace{0.2em} ^{g}\textrm{T}_c \hspace{0.2em} ^{c}{\textrm{T}_t}^{(2)} \\ (^{b}{\textrm{T}_g}^{(2)})^{-1} \hspace{0.2em} ^{b}{\textrm{T}_g}^{(1)} \hspace{0.2em} ^{g}\textrm{T}_c &= \hspace{0.1em} ^{g}\textrm{T}_c \hspace{0.2em} ^{c}{\textrm{T}_t}^{(2)} (^{c}{\textrm{T}_t}^{(1)})^{-1} \\ \textrm{A}_i \textrm{X} &= \textrm{X} \textrm{B}_i \\ \end{align*} \f]

  • for an eye-to-hand configuration \f[ \begin{align*} ^{g}{\textrm{T}_b}^{(1)} \hspace{0.2em} ^{b}\textrm{T}_c \hspace{0.2em} ^{c}{\textrm{T}_t}^{(1)} &= \hspace{0.1em} ^{g}{\textrm{T}_b}^{(2)} \hspace{0.2em} ^{b}\textrm{T}_c \hspace{0.2em} ^{c}{\textrm{T}_t}^{(2)} \\ (^{g}{\textrm{T}_b}^{(2)})^{-1} \hspace{0.2em} ^{g}{\textrm{T}_b}^{(1)} \hspace{0.2em} ^{b}\textrm{T}_c &= \hspace{0.1em} ^{b}\textrm{T}_c \hspace{0.2em} ^{c}{\textrm{T}_t}^{(2)} (^{c}{\textrm{T}_t}^{(1)})^{-1} \\ \textrm{A}_i \textrm{X} &= \textrm{X} \textrm{B}_i \\ \end{align*} \f]

\note Additional information can be found on this website. \note A minimum of 2 motions with non parallel rotation axes are necessary to determine the hand-eye transformation. So at least 3 different poses are required, but it is strongly recommended to use many more poses.

Python prototype (for reference only):

calibrateHandEye(R_gripper2base, t_gripper2base, R_target2cam, t_target2cam[, R_cam2gripper[, t_cam2gripper[, method]]]) -> R_cam2gripper, t_cam2gripper
Link to this function

calibrateRobotWorldHandEye(r_world2cam, t_world2cam, r_base2gripper, t_base2gripper)

View Source

Computes Robot-World/Hand-Eye calibration: \f$_{}^{w}\textrm{T}_b\f$ and \f$_{}^{c}\textrm{T}_g\f$

Positional Arguments
  • r_world2cam: [Evision.Mat].

    Rotation part extracted from the homogeneous matrix that transforms a point expressed in the world frame to the camera frame (\f$_{}^{c}\textrm{T}_w\f$). This is a vector (vector<Mat>) that contains the rotation, (3x3) rotation matrices or (3x1) rotation vectors, for all the transformations from world frame to the camera frame.

  • t_world2cam: [Evision.Mat].

    Translation part extracted from the homogeneous matrix that transforms a point expressed in the world frame to the camera frame (\f$_{}^{c}\textrm{T}_w\f$). This is a vector (vector<Mat>) that contains the (3x1) translation vectors for all the transformations from world frame to the camera frame.

  • r_base2gripper: [Evision.Mat].

    Rotation part extracted from the homogeneous matrix that transforms a point expressed in the robot base frame to the gripper frame (\f$_{}^{g}\textrm{T}_b\f$). This is a vector (vector<Mat>) that contains the rotation, (3x3) rotation matrices or (3x1) rotation vectors, for all the transformations from robot base frame to the gripper frame.

  • t_base2gripper: [Evision.Mat].

    Rotation part extracted from the homogeneous matrix that transforms a point expressed in the robot base frame to the gripper frame (\f$_{}^{g}\textrm{T}_b\f$). This is a vector (vector<Mat>) that contains the (3x1) translation vectors for all the transformations from robot base frame to the gripper frame.

Keyword Arguments
  • method: RobotWorldHandEyeCalibrationMethod.

    One of the implemented Robot-World/Hand-Eye calibration method, see cv::RobotWorldHandEyeCalibrationMethod

Return
  • r_base2world: Evision.Mat.t().

    Estimated (3x3) rotation part extracted from the homogeneous matrix that transforms a point expressed in the robot base frame to the world frame (\f$_{}^{w}\textrm{T}_b\f$).

  • t_base2world: Evision.Mat.t().

    Estimated (3x1) translation part extracted from the homogeneous matrix that transforms a point expressed in the robot base frame to the world frame (\f$_{}^{w}\textrm{T}_b\f$).

  • r_gripper2cam: Evision.Mat.t().

    Estimated (3x3) rotation part extracted from the homogeneous matrix that transforms a point expressed in the gripper frame to the camera frame (\f$_{}^{c}\textrm{T}_g\f$).

  • t_gripper2cam: Evision.Mat.t().

    Estimated (3x1) translation part extracted from the homogeneous matrix that transforms a point expressed in the gripper frame to the camera frame (\f$_{}^{c}\textrm{T}_g\f$).

The function performs the Robot-World/Hand-Eye calibration using various methods. One approach consists in estimating the rotation then the translation (separable solutions):

  • M. Shah, Solving the robot-world/hand-eye calibration problem using the kronecker product \cite Shah2013SolvingTR

Another approach consists in estimating simultaneously the rotation and the translation (simultaneous solutions), with the following implemented method:

  • A. Li, L. Wang, and D. Wu, Simultaneous robot-world and hand-eye calibration using dual-quaternions and kronecker product \cite Li2010SimultaneousRA

The following picture describes the Robot-World/Hand-Eye calibration problem where the transformations between a robot and a world frame and between a robot gripper ("hand") and a camera ("eye") mounted at the robot end-effector have to be estimated. The calibration procedure is the following:

  • a static calibration pattern is used to estimate the transformation between the target frame and the camera frame

  • the robot gripper is moved in order to acquire several poses

  • for each pose, the homogeneous transformation between the gripper frame and the robot base frame is recorded using for instance the robot kinematics \f[ \begin{bmatrix} X_g\\ Y_g\\ Z_g\\ 1 \end{bmatrix}= \begin{bmatrix} _{}^{g}\textrm{R}_b & _{}^{g}\textrm{t}_b \\ 0_{1 \times 3} & 1 \end{bmatrix} \begin{bmatrix} X_b\\ Y_b\\ Z_b\\ 1 \end{bmatrix} \f]

  • for each pose, the homogeneous transformation between the calibration target frame (the world frame) and the camera frame is recorded using for instance a pose estimation method (PnP) from 2D-3D point correspondences \f[ \begin{bmatrix} X_c\\ Y_c\\ Z_c\\ 1 \end{bmatrix}= \begin{bmatrix} _{}^{c}\textrm{R}_w & _{}^{c}\textrm{t}_w \\ 0_{1 \times 3} & 1 \end{bmatrix} \begin{bmatrix} X_w\\ Y_w\\ Z_w\\ 1 \end{bmatrix} \f]

The Robot-World/Hand-Eye calibration procedure returns the following homogeneous transformations \f[ \begin{bmatrix} X_w\\ Y_w\\ Z_w\\ 1 \end{bmatrix} \begin{bmatrix} _{}^{w}\textrm{R}_b & _{}^{w}\textrm{t}_b \\ 0_{1 \times 3} & 1 \end{bmatrix} \begin{bmatrix} X_b\\ Y_b\\ Z_b\\ 1 \end{bmatrix} \f] \f[ \begin{bmatrix} X_c\\ Y_c\\ Z_c\\ 1 \end{bmatrix} \begin{bmatrix} _{}^{c}\textrm{R}_g & _{}^{c}\textrm{t}_g \\ 0_{1 \times 3} & 1 \end{bmatrix} \begin{bmatrix} X_g\\ Y_g\\ Z_g\\ 1 \end{bmatrix} \f] This problem is also known as solving the \f$\mathbf{A}\mathbf{X}=\mathbf{Z}\mathbf{B}\f$ equation, with:

  • \f$\mathbf{A} \Leftrightarrow \hspace{0.1em} _{}^{c}\textrm{T}_w\f$
  • \f$\mathbf{X} \Leftrightarrow \hspace{0.1em} _{}^{w}\textrm{T}_b\f$
  • \f$\mathbf{Z} \Leftrightarrow \hspace{0.1em} _{}^{c}\textrm{T}_g\f$
  • \f$\mathbf{B} \Leftrightarrow \hspace{0.1em} _{}^{g}\textrm{T}_b\f$

\note At least 3 measurements are required (input vectors size must be greater or equal to 3).

Python prototype (for reference only):

calibrateRobotWorldHandEye(R_world2cam, t_world2cam, R_base2gripper, t_base2gripper[, R_base2world[, t_base2world[, R_gripper2cam[, t_gripper2cam[, method]]]]]) -> R_base2world, t_base2world, R_gripper2cam, t_gripper2cam
Link to this function

calibrateRobotWorldHandEye(r_world2cam, t_world2cam, r_base2gripper, t_base2gripper, opts)

View Source
@spec calibrateRobotWorldHandEye(
  [Evision.Mat.maybe_mat_in()],
  [Evision.Mat.maybe_mat_in()],
  [Evision.Mat.maybe_mat_in()],
  [Evision.Mat.maybe_mat_in()],
  [{atom(), term()}, ...] | nil
) ::
  {Evision.Mat.t(), Evision.Mat.t(), Evision.Mat.t(), Evision.Mat.t()}
  | {:error, String.t()}

Computes Robot-World/Hand-Eye calibration: \f$_{}^{w}\textrm{T}_b\f$ and \f$_{}^{c}\textrm{T}_g\f$

Positional Arguments
  • r_world2cam: [Evision.Mat].

    Rotation part extracted from the homogeneous matrix that transforms a point expressed in the world frame to the camera frame (\f$_{}^{c}\textrm{T}_w\f$). This is a vector (vector<Mat>) that contains the rotation, (3x3) rotation matrices or (3x1) rotation vectors, for all the transformations from world frame to the camera frame.

  • t_world2cam: [Evision.Mat].

    Translation part extracted from the homogeneous matrix that transforms a point expressed in the world frame to the camera frame (\f$_{}^{c}\textrm{T}_w\f$). This is a vector (vector<Mat>) that contains the (3x1) translation vectors for all the transformations from world frame to the camera frame.

  • r_base2gripper: [Evision.Mat].

    Rotation part extracted from the homogeneous matrix that transforms a point expressed in the robot base frame to the gripper frame (\f$_{}^{g}\textrm{T}_b\f$). This is a vector (vector<Mat>) that contains the rotation, (3x3) rotation matrices or (3x1) rotation vectors, for all the transformations from robot base frame to the gripper frame.

  • t_base2gripper: [Evision.Mat].

    Rotation part extracted from the homogeneous matrix that transforms a point expressed in the robot base frame to the gripper frame (\f$_{}^{g}\textrm{T}_b\f$). This is a vector (vector<Mat>) that contains the (3x1) translation vectors for all the transformations from robot base frame to the gripper frame.

Keyword Arguments
  • method: RobotWorldHandEyeCalibrationMethod.

    One of the implemented Robot-World/Hand-Eye calibration method, see cv::RobotWorldHandEyeCalibrationMethod

Return
  • r_base2world: Evision.Mat.t().

    Estimated (3x3) rotation part extracted from the homogeneous matrix that transforms a point expressed in the robot base frame to the world frame (\f$_{}^{w}\textrm{T}_b\f$).

  • t_base2world: Evision.Mat.t().

    Estimated (3x1) translation part extracted from the homogeneous matrix that transforms a point expressed in the robot base frame to the world frame (\f$_{}^{w}\textrm{T}_b\f$).

  • r_gripper2cam: Evision.Mat.t().

    Estimated (3x3) rotation part extracted from the homogeneous matrix that transforms a point expressed in the gripper frame to the camera frame (\f$_{}^{c}\textrm{T}_g\f$).

  • t_gripper2cam: Evision.Mat.t().

    Estimated (3x1) translation part extracted from the homogeneous matrix that transforms a point expressed in the gripper frame to the camera frame (\f$_{}^{c}\textrm{T}_g\f$).

The function performs the Robot-World/Hand-Eye calibration using various methods. One approach consists in estimating the rotation then the translation (separable solutions):

  • M. Shah, Solving the robot-world/hand-eye calibration problem using the kronecker product \cite Shah2013SolvingTR

Another approach consists in estimating simultaneously the rotation and the translation (simultaneous solutions), with the following implemented method:

  • A. Li, L. Wang, and D. Wu, Simultaneous robot-world and hand-eye calibration using dual-quaternions and kronecker product \cite Li2010SimultaneousRA

The following picture describes the Robot-World/Hand-Eye calibration problem where the transformations between a robot and a world frame and between a robot gripper ("hand") and a camera ("eye") mounted at the robot end-effector have to be estimated. The calibration procedure is the following:

  • a static calibration pattern is used to estimate the transformation between the target frame and the camera frame

  • the robot gripper is moved in order to acquire several poses

  • for each pose, the homogeneous transformation between the gripper frame and the robot base frame is recorded using for instance the robot kinematics \f[ \begin{bmatrix} X_g\\ Y_g\\ Z_g\\ 1 \end{bmatrix}= \begin{bmatrix} _{}^{g}\textrm{R}_b & _{}^{g}\textrm{t}_b \\ 0_{1 \times 3} & 1 \end{bmatrix} \begin{bmatrix} X_b\\ Y_b\\ Z_b\\ 1 \end{bmatrix} \f]

  • for each pose, the homogeneous transformation between the calibration target frame (the world frame) and the camera frame is recorded using for instance a pose estimation method (PnP) from 2D-3D point correspondences \f[ \begin{bmatrix} X_c\\ Y_c\\ Z_c\\ 1 \end{bmatrix}= \begin{bmatrix} _{}^{c}\textrm{R}_w & _{}^{c}\textrm{t}_w \\ 0_{1 \times 3} & 1 \end{bmatrix} \begin{bmatrix} X_w\\ Y_w\\ Z_w\\ 1 \end{bmatrix} \f]

The Robot-World/Hand-Eye calibration procedure returns the following homogeneous transformations \f[ \begin{bmatrix} X_w\\ Y_w\\ Z_w\\ 1 \end{bmatrix} \begin{bmatrix} _{}^{w}\textrm{R}_b & _{}^{w}\textrm{t}_b \\ 0_{1 \times 3} & 1 \end{bmatrix} \begin{bmatrix} X_b\\ Y_b\\ Z_b\\ 1 \end{bmatrix} \f] \f[ \begin{bmatrix} X_c\\ Y_c\\ Z_c\\ 1 \end{bmatrix} \begin{bmatrix} _{}^{c}\textrm{R}_g & _{}^{c}\textrm{t}_g \\ 0_{1 \times 3} & 1 \end{bmatrix} \begin{bmatrix} X_g\\ Y_g\\ Z_g\\ 1 \end{bmatrix} \f] This problem is also known as solving the \f$\mathbf{A}\mathbf{X}=\mathbf{Z}\mathbf{B}\f$ equation, with:

  • \f$\mathbf{A} \Leftrightarrow \hspace{0.1em} _{}^{c}\textrm{T}_w\f$
  • \f$\mathbf{X} \Leftrightarrow \hspace{0.1em} _{}^{w}\textrm{T}_b\f$
  • \f$\mathbf{Z} \Leftrightarrow \hspace{0.1em} _{}^{c}\textrm{T}_g\f$
  • \f$\mathbf{B} \Leftrightarrow \hspace{0.1em} _{}^{g}\textrm{T}_b\f$

\note At least 3 measurements are required (input vectors size must be greater or equal to 3).

Python prototype (for reference only):

calibrateRobotWorldHandEye(R_world2cam, t_world2cam, R_base2gripper, t_base2gripper[, R_base2world[, t_base2world[, R_gripper2cam[, t_gripper2cam[, method]]]]]) -> R_base2world, t_base2world, R_gripper2cam, t_gripper2cam
Link to this function

calibrationMatrixValues(cameraMatrix, imageSize, apertureWidth, apertureHeight)

View Source
@spec calibrationMatrixValues(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  number(),
  number()
) ::
  {number(), number(), number(), {number(), number()}, number()}
  | {:error, String.t()}

Computes useful camera characteristics from the camera intrinsic matrix.

Positional Arguments
  • cameraMatrix: Evision.Mat.t().

    Input camera intrinsic matrix that can be estimated by #calibrateCamera or #stereoCalibrate .

  • imageSize: Size.

    Input image size in pixels.

  • apertureWidth: double.

    Physical width in mm of the sensor.

  • apertureHeight: double.

    Physical height in mm of the sensor.

Return
  • fovx: double.

    Output field of view in degrees along the horizontal sensor axis.

  • fovy: double.

    Output field of view in degrees along the vertical sensor axis.

  • focalLength: double.

    Focal length of the lens in mm.

  • principalPoint: Point2d.

    Principal point in mm.

  • aspectRatio: double.

    \f$f_y/f_x\f$

The function computes various useful camera characteristics from the previously estimated camera matrix. Note: Do keep in mind that the unity measure 'mm' stands for whatever unit of measure one chooses for the chessboard pitch (it can thus be any value).

Python prototype (for reference only):

calibrationMatrixValues(cameraMatrix, imageSize, apertureWidth, apertureHeight) -> fovx, fovy, focalLength, principalPoint, aspectRatio
Link to this function

camShift(probImage, window, criteria)

View Source
@spec camShift(
  Evision.Mat.maybe_mat_in(),
  {number(), number(), number(), number()},
  {integer(), integer(), number()}
) ::
  {Evision.RotatedRect.t(), {number(), number(), number(), number()}}
  | {:error, String.t()}

Finds an object center, size, and orientation.

Positional Arguments
  • probImage: Evision.Mat.t().

    Back projection of the object histogram. See calcBackProject.

  • criteria: TermCriteria.

    Stop criteria for the underlying meanShift. returns (in old interfaces) Number of iterations CAMSHIFT took to converge The function implements the CAMSHIFT object tracking algorithm @cite Bradski98 . First, it finds an object center using meanShift and then adjusts the window size and finds the optimal rotation. The function returns the rotated rectangle structure that includes the object position, size, and orientation. The next position of the search window can be obtained with RotatedRect::boundingRect()

Return
  • retval: {centre={x, y}, size={s1, s2}, angle}

  • window: Rect.

    Initial search window.

See the OpenCV sample camshiftdemo.c that tracks colored objects. Note:

  • (Python) A sample explaining the camshift tracking algorithm can be found at opencv_source_code/samples/python/camshift.py

Python prototype (for reference only):

CamShift(probImage, window, criteria) -> retval, window
Link to this function

canny(image, threshold1, threshold2)

View Source
@spec canny(Evision.Mat.maybe_mat_in(), number(), number()) ::
  Evision.Mat.t() | {:error, String.t()}

Finds edges in an image using the Canny algorithm @cite Canny86 .

Positional Arguments
  • image: Evision.Mat.t().

    8-bit input image.

  • threshold1: double.

    first threshold for the hysteresis procedure.

  • threshold2: double.

    second threshold for the hysteresis procedure.

Keyword Arguments
  • apertureSize: int.

    aperture size for the Sobel operator.

  • l2gradient: bool.

    a flag, indicating whether a more accurate \f$L_2\f$ norm \f$=\sqrt{(dI/dx)^2 + (dI/dy)^2}\f$ should be used to calculate the image gradient magnitude ( L2gradient=true ), or whether the default \f$L_1\f$ norm \f$=|dI/dx|+|dI/dy|\f$ is enough ( L2gradient=false ).

Return
  • edges: Evision.Mat.t().

    output edge map; single channels 8-bit image, which has the same size as image .

The function finds edges in the input image and marks them in the output map edges using the Canny algorithm. The smallest value between threshold1 and threshold2 is used for edge linking. The largest value is used to find initial segments of strong edges. See http://en.wikipedia.org/wiki/Canny_edge_detector

Python prototype (for reference only):

Canny(image, threshold1, threshold2[, edges[, apertureSize[, L2gradient]]]) -> edges
Link to this function

canny(image, threshold1, threshold2, opts)

View Source
@spec canny(
  Evision.Mat.maybe_mat_in(),
  number(),
  number(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}
@spec canny(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  number(),
  number()
) ::
  Evision.Mat.t() | {:error, String.t()}

Variant 1:

Canny

Positional Arguments
  • dx: Evision.Mat.t().

    16-bit x derivative of input image (CV_16SC1 or CV_16SC3).

  • dy: Evision.Mat.t().

    16-bit y derivative of input image (same type as dx).

  • threshold1: double.

    first threshold for the hysteresis procedure.

  • threshold2: double.

    second threshold for the hysteresis procedure.

Keyword Arguments
  • l2gradient: bool.

    a flag, indicating whether a more accurate \f$L_2\f$ norm \f$=\sqrt{(dI/dx)^2 + (dI/dy)^2}\f$ should be used to calculate the image gradient magnitude ( L2gradient=true ), or whether the default \f$L_1\f$ norm \f$=|dI/dx|+|dI/dy|\f$ is enough ( L2gradient=false ).

Return
  • edges: Evision.Mat.t().

    output edge map; single channels 8-bit image, which has the same size as image .

Finds edges in an image using the Canny algorithm with custom image gradient.

Python prototype (for reference only):

Canny(dx, dy, threshold1, threshold2[, edges[, L2gradient]]) -> edges

Variant 2:

Finds edges in an image using the Canny algorithm @cite Canny86 .

Positional Arguments
  • image: Evision.Mat.t().

    8-bit input image.

  • threshold1: double.

    first threshold for the hysteresis procedure.

  • threshold2: double.

    second threshold for the hysteresis procedure.

Keyword Arguments
  • apertureSize: int.

    aperture size for the Sobel operator.

  • l2gradient: bool.

    a flag, indicating whether a more accurate \f$L_2\f$ norm \f$=\sqrt{(dI/dx)^2 + (dI/dy)^2}\f$ should be used to calculate the image gradient magnitude ( L2gradient=true ), or whether the default \f$L_1\f$ norm \f$=|dI/dx|+|dI/dy|\f$ is enough ( L2gradient=false ).

Return
  • edges: Evision.Mat.t().

    output edge map; single channels 8-bit image, which has the same size as image .

The function finds edges in the input image and marks them in the output map edges using the Canny algorithm. The smallest value between threshold1 and threshold2 is used for edge linking. The largest value is used to find initial segments of strong edges. See http://en.wikipedia.org/wiki/Canny_edge_detector

Python prototype (for reference only):

Canny(image, threshold1, threshold2[, edges[, apertureSize[, L2gradient]]]) -> edges
Link to this function

canny(dx, dy, threshold1, threshold2, opts)

View Source
@spec canny(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  number(),
  number(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Canny

Positional Arguments
  • dx: Evision.Mat.t().

    16-bit x derivative of input image (CV_16SC1 or CV_16SC3).

  • dy: Evision.Mat.t().

    16-bit y derivative of input image (same type as dx).

  • threshold1: double.

    first threshold for the hysteresis procedure.

  • threshold2: double.

    second threshold for the hysteresis procedure.

Keyword Arguments
  • l2gradient: bool.

    a flag, indicating whether a more accurate \f$L_2\f$ norm \f$=\sqrt{(dI/dx)^2 + (dI/dy)^2}\f$ should be used to calculate the image gradient magnitude ( L2gradient=true ), or whether the default \f$L_1\f$ norm \f$=|dI/dx|+|dI/dy|\f$ is enough ( L2gradient=false ).

Return
  • edges: Evision.Mat.t().

    output edge map; single channels 8-bit image, which has the same size as image .

Finds edges in an image using the Canny algorithm with custom image gradient.

Python prototype (for reference only):

Canny(dx, dy, threshold1, threshold2[, edges[, L2gradient]]) -> edges

Calculates the magnitude and angle of 2D vectors.

Positional Arguments
  • x: Evision.Mat.t().

    array of x-coordinates; this must be a single-precision or double-precision floating-point array.

  • y: Evision.Mat.t().

    array of y-coordinates, that must have the same size and same type as x.

Keyword Arguments
  • angleInDegrees: bool.

    a flag, indicating whether the angles are measured in radians (which is by default), or in degrees.

Return
  • magnitude: Evision.Mat.t().

    output array of magnitudes of the same size and type as x.

  • angle: Evision.Mat.t().

    output array of angles that has the same size and type as x; the angles are measured in radians (from 0 to 2*Pi) or in degrees (0 to 360 degrees).

The function cv::cartToPolar calculates either the magnitude, angle, or both for every 2D vector (x(I),y(I)): \f[\begin{array}{l} \texttt{magnitude} (I)= \sqrt{\texttt{x}(I)^2+\texttt{y}(I)^2} , \\ \texttt{angle} (I)= \texttt{atan2} ( \texttt{y} (I), \texttt{x} (I))[ \cdot180 / \pi ] \end{array}\f] The angles are calculated with accuracy about 0.3 degrees. For the point (0,0), the angle is set to 0. @sa Sobel, Scharr

Python prototype (for reference only):

cartToPolar(x, y[, magnitude[, angle[, angleInDegrees]]]) -> magnitude, angle
@spec cartToPolar(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Calculates the magnitude and angle of 2D vectors.

Positional Arguments
  • x: Evision.Mat.t().

    array of x-coordinates; this must be a single-precision or double-precision floating-point array.

  • y: Evision.Mat.t().

    array of y-coordinates, that must have the same size and same type as x.

Keyword Arguments
  • angleInDegrees: bool.

    a flag, indicating whether the angles are measured in radians (which is by default), or in degrees.

Return
  • magnitude: Evision.Mat.t().

    output array of magnitudes of the same size and type as x.

  • angle: Evision.Mat.t().

    output array of angles that has the same size and type as x; the angles are measured in radians (from 0 to 2*Pi) or in degrees (0 to 360 degrees).

The function cv::cartToPolar calculates either the magnitude, angle, or both for every 2D vector (x(I),y(I)): \f[\begin{array}{l} \texttt{magnitude} (I)= \sqrt{\texttt{x}(I)^2+\texttt{y}(I)^2} , \\ \texttt{angle} (I)= \texttt{atan2} ( \texttt{y} (I), \texttt{x} (I))[ \cdot180 / \pi ] \end{array}\f] The angles are calculated with accuracy about 0.3 degrees. For the point (0,0), the angle is set to 0. @sa Sobel, Scharr

Python prototype (for reference only):

cartToPolar(x, y[, magnitude[, angle[, angleInDegrees]]]) -> magnitude, angle
Link to this function

checkChessboard(img, size)

View Source
@spec checkChessboard(
  Evision.Mat.maybe_mat_in(),
  {number(), number()}
) :: boolean() | {:error, String.t()}

checkChessboard

Positional Arguments
  • img: Evision.Mat.t()
  • size: Size
Return
  • retval: bool

Python prototype (for reference only):

checkChessboard(img, size) -> retval
Link to this function

checkHardwareSupport(feature)

View Source
@spec checkHardwareSupport(integer()) :: boolean() | {:error, String.t()}

Returns true if the specified feature is supported by the host hardware.

Positional Arguments
  • feature: int.

    The feature of interest, one of cv::CpuFeatures

Return
  • retval: bool

The function returns true if the host hardware supports the specified feature. When user calls setUseOptimized(false), the subsequent calls to checkHardwareSupport() will return false until setUseOptimized(true) is called. This way user can dynamically switch on and off the optimized code in OpenCV.

Python prototype (for reference only):

checkHardwareSupport(feature) -> retval
@spec checkRange(Evision.Mat.maybe_mat_in()) ::
  {number(), number()} | false | {:error, String.t()}

Checks every element of an input array for invalid values.

Positional Arguments
  • a: Evision.Mat.t().

    input array.

Keyword Arguments
  • quiet: bool.

    a flag, indicating whether the functions quietly return false when the array elements are out of range or they throw an exception.

  • minVal: double.

    inclusive lower boundary of valid values range.

  • maxVal: double.

    exclusive upper boundary of valid values range.

Return
  • retval: bool

  • pos: Point*.

    optional output parameter, when not NULL, must be a pointer to array of src.dims elements.

The function cv::checkRange checks that every array element is neither NaN nor infinite. When minVal >

  • DBL_MAX and maxVal \< DBL_MAX, the function also checks that each value is between minVal and maxVal. In case of multi-channel arrays, each channel is processed independently. If some values are out of range, position of the first outlier is stored in pos (when pos != NULL). Then, the function either returns false (when quiet=true) or throws an exception.

Python prototype (for reference only):

checkRange(a[, quiet[, minVal[, maxVal]]]) -> retval, pos
@spec checkRange(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  {number(), number()} | false | {:error, String.t()}

Checks every element of an input array for invalid values.

Positional Arguments
  • a: Evision.Mat.t().

    input array.

Keyword Arguments
  • quiet: bool.

    a flag, indicating whether the functions quietly return false when the array elements are out of range or they throw an exception.

  • minVal: double.

    inclusive lower boundary of valid values range.

  • maxVal: double.

    exclusive upper boundary of valid values range.

Return
  • retval: bool

  • pos: Point*.

    optional output parameter, when not NULL, must be a pointer to array of src.dims elements.

The function cv::checkRange checks that every array element is neither NaN nor infinite. When minVal >

  • DBL_MAX and maxVal \< DBL_MAX, the function also checks that each value is between minVal and maxVal. In case of multi-channel arrays, each channel is processed independently. If some values are out of range, position of the first outlier is stored in pos (when pos != NULL). Then, the function either returns false (when quiet=true) or throws an exception.

Python prototype (for reference only):

checkRange(a[, quiet[, minVal[, maxVal]]]) -> retval, pos
Link to this function

circle(img, center, radius, color)

View Source
@spec circle(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  integer(),
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()}
) :: Evision.Mat.t() | {:error, String.t()}

Draws a circle.

Positional Arguments
  • center: Point.

    Center of the circle.

  • radius: int.

    Radius of the circle.

  • color: Scalar.

    Circle color.

Keyword Arguments
  • thickness: int.

    Thickness of the circle outline, if positive. Negative values, like #FILLED, mean that a filled circle is to be drawn.

  • lineType: int.

    Type of the circle boundary. See #LineTypes

  • shift: int.

    Number of fractional bits in the coordinates of the center and in the radius value.

Return
  • img: Evision.Mat.t().

    Image where the circle is drawn.

The function cv::circle draws a simple or filled circle with a given center and radius.

Python prototype (for reference only):

circle(img, center, radius, color[, thickness[, lineType[, shift]]]) -> img
Link to this function

circle(img, center, radius, color, opts)

View Source
@spec circle(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  integer(),
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()},
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Draws a circle.

Positional Arguments
  • center: Point.

    Center of the circle.

  • radius: int.

    Radius of the circle.

  • color: Scalar.

    Circle color.

Keyword Arguments
  • thickness: int.

    Thickness of the circle outline, if positive. Negative values, like #FILLED, mean that a filled circle is to be drawn.

  • lineType: int.

    Type of the circle boundary. See #LineTypes

  • shift: int.

    Number of fractional bits in the coordinates of the center and in the radius value.

Return
  • img: Evision.Mat.t().

    Image where the circle is drawn.

The function cv::circle draws a simple or filled circle with a given center and radius.

Python prototype (for reference only):

circle(img, center, radius, color[, thickness[, lineType[, shift]]]) -> img
Link to this function

clipLine(imgRect, pt1, pt2)

View Source
@spec clipLine(
  {number(), number(), number(), number()},
  {number(), number()},
  {number(), number()}
) ::
  {{number(), number()}, {number(), number()}} | false | {:error, String.t()}

clipLine

Positional Arguments
  • imgRect: Rect.

    Image rectangle.

Return
  • retval: bool

  • pt1: Point.

    First line point.

  • pt2: Point.

    Second line point.

Has overloading in C++

Python prototype (for reference only):

clipLine(imgRect, pt1, pt2) -> retval, pt1, pt2
@spec colorChange(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

Given an original color image, two differently colored versions of this image can be mixed seamlessly.

Positional Arguments
  • src: Evision.Mat.t().

    Input 8-bit 3-channel image.

  • mask: Evision.Mat.t().

    Input 8-bit 1 or 3-channel image.

Keyword Arguments
  • red_mul: float.

    R-channel multiply factor.

  • green_mul: float.

    G-channel multiply factor.

  • blue_mul: float.

    B-channel multiply factor.

Return
  • dst: Evision.Mat.t().

    Output image with the same size and type as src .

Multiplication factor is between .5 to 2.5.

Python prototype (for reference only):

colorChange(src, mask[, dst[, red_mul[, green_mul[, blue_mul]]]]) -> dst
Link to this function

colorChange(src, mask, opts)

View Source
@spec colorChange(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Given an original color image, two differently colored versions of this image can be mixed seamlessly.

Positional Arguments
  • src: Evision.Mat.t().

    Input 8-bit 3-channel image.

  • mask: Evision.Mat.t().

    Input 8-bit 1 or 3-channel image.

Keyword Arguments
  • red_mul: float.

    R-channel multiply factor.

  • green_mul: float.

    G-channel multiply factor.

  • blue_mul: float.

    B-channel multiply factor.

Return
  • dst: Evision.Mat.t().

    Output image with the same size and type as src .

Multiplication factor is between .5 to 2.5.

Python prototype (for reference only):

colorChange(src, mask[, dst[, red_mul[, green_mul[, blue_mul]]]]) -> dst
Link to this function

compare(src1, src2, cmpop)

View Source

Performs the per-element comparison of two arrays or an array and scalar value.

Positional Arguments
  • src1: Evision.Mat.t().

    first input array or a scalar; when it is an array, it must have a single channel.

  • src2: Evision.Mat.t().

    second input array or a scalar; when it is an array, it must have a single channel.

  • cmpop: int.

    a flag, that specifies correspondence between the arrays (cv::CmpTypes)

Return
  • dst: Evision.Mat.t().

    output array of type ref CV_8U that has the same size and the same number of channels as the input arrays.

The function compares: Elements of two arrays when src1 and src2 have the same size: \f[\texttt{dst} (I) = \texttt{src1} (I) \,\texttt{cmpop}\, \texttt{src2} (I)\f] Elements of src1 with a scalar src2 when src2 is constructed from Scalar or has a single element: \f[\texttt{dst} (I) = \texttt{src1}(I) \,\texttt{cmpop}\, \texttt{src2}\f] src1 with elements of src2 when src1 is constructed from Scalar or has a single element: \f[\texttt{dst} (I) = \texttt{src1} \,\texttt{cmpop}\, \texttt{src2} (I)\f] When the comparison result is true, the corresponding element of output array is set to 255. The comparison operations can be replaced with the equivalent matrix expressions:

Mat dst1 = src1 >= src2;
Mat dst2 = src1 < 8;
...

@sa checkRange, min, max, threshold

Python prototype (for reference only):

compare(src1, src2, cmpop[, dst]) -> dst
Link to this function

compare(src1, src2, cmpop, opts)

View Source
@spec compare(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  integer(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Performs the per-element comparison of two arrays or an array and scalar value.

Positional Arguments
  • src1: Evision.Mat.t().

    first input array or a scalar; when it is an array, it must have a single channel.

  • src2: Evision.Mat.t().

    second input array or a scalar; when it is an array, it must have a single channel.

  • cmpop: int.

    a flag, that specifies correspondence between the arrays (cv::CmpTypes)

Return
  • dst: Evision.Mat.t().

    output array of type ref CV_8U that has the same size and the same number of channels as the input arrays.

The function compares: Elements of two arrays when src1 and src2 have the same size: \f[\texttt{dst} (I) = \texttt{src1} (I) \,\texttt{cmpop}\, \texttt{src2} (I)\f] Elements of src1 with a scalar src2 when src2 is constructed from Scalar or has a single element: \f[\texttt{dst} (I) = \texttt{src1}(I) \,\texttt{cmpop}\, \texttt{src2}\f] src1 with elements of src2 when src1 is constructed from Scalar or has a single element: \f[\texttt{dst} (I) = \texttt{src1} \,\texttt{cmpop}\, \texttt{src2} (I)\f] When the comparison result is true, the corresponding element of output array is set to 255. The comparison operations can be replaced with the equivalent matrix expressions:

Mat dst1 = src1 >= src2;
Mat dst2 = src1 < 8;
...

@sa checkRange, min, max, threshold

Python prototype (for reference only):

compare(src1, src2, cmpop[, dst]) -> dst
Link to this function

compareHist(h1, h2, method)

View Source
@spec compareHist(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in(), integer()) ::
  number() | {:error, String.t()}

Compares two histograms.

Positional Arguments
  • h1: Evision.Mat.t().

    First compared histogram.

  • h2: Evision.Mat.t().

    Second compared histogram of the same size as H1 .

  • method: int.

    Comparison method, see #HistCompMethods

Return
  • retval: double

The function cv::compareHist compares two dense or two sparse histograms using the specified method. The function returns \f$d(H_1, H_2)\f$ . While the function works well with 1-, 2-, 3-dimensional dense histograms, it may not be suitable for high-dimensional sparse histograms. In such histograms, because of aliasing and sampling problems, the coordinates of non-zero histogram bins can slightly shift. To compare such histograms or more general sparse configurations of weighted points, consider using the #EMD function.

Python prototype (for reference only):

compareHist(H1, H2, method) -> retval
@spec completeSymm(Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

Copies the lower or the upper half of a square matrix to its another half.

Keyword Arguments
  • lowerToUpper: bool.

    operation flag; if true, the lower half is copied to the upper half. Otherwise, the upper half is copied to the lower half.

Return
  • m: Evision.Mat.t().

    input-output floating-point square matrix.

The function cv::completeSymm copies the lower or the upper half of a square matrix to its another half. The matrix diagonal remains unchanged:

  • \f$\texttt{m}_{ij}=\texttt{m}_{ji}\f$ for \f$i > j\f$ if lowerToUpper=false

  • \f$\texttt{m}_{ij}=\texttt{m}_{ji}\f$ for \f$i < j\f$ if lowerToUpper=true

@sa flip, transpose

Python prototype (for reference only):

completeSymm(m[, lowerToUpper]) -> m
@spec completeSymm(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Copies the lower or the upper half of a square matrix to its another half.

Keyword Arguments
  • lowerToUpper: bool.

    operation flag; if true, the lower half is copied to the upper half. Otherwise, the upper half is copied to the lower half.

Return
  • m: Evision.Mat.t().

    input-output floating-point square matrix.

The function cv::completeSymm copies the lower or the upper half of a square matrix to its another half. The matrix diagonal remains unchanged:

  • \f$\texttt{m}_{ij}=\texttt{m}_{ji}\f$ for \f$i > j\f$ if lowerToUpper=false

  • \f$\texttt{m}_{ij}=\texttt{m}_{ji}\f$ for \f$i < j\f$ if lowerToUpper=true

@sa flip, transpose

Python prototype (for reference only):

completeSymm(m[, lowerToUpper]) -> m
Link to this function

composeRT(rvec1, tvec1, rvec2, tvec2)

View Source

Combines two rotation-and-shift transformations.

Positional Arguments
  • rvec1: Evision.Mat.t().

    First rotation vector.

  • tvec1: Evision.Mat.t().

    First translation vector.

  • rvec2: Evision.Mat.t().

    Second rotation vector.

  • tvec2: Evision.Mat.t().

    Second translation vector.

Return
  • rvec3: Evision.Mat.t().

    Output rotation vector of the superposition.

  • tvec3: Evision.Mat.t().

    Output translation vector of the superposition.

  • dr3dr1: Evision.Mat.t().

    Optional output derivative of rvec3 with regard to rvec1

  • dr3dt1: Evision.Mat.t().

    Optional output derivative of rvec3 with regard to tvec1

  • dr3dr2: Evision.Mat.t().

    Optional output derivative of rvec3 with regard to rvec2

  • dr3dt2: Evision.Mat.t().

    Optional output derivative of rvec3 with regard to tvec2

  • dt3dr1: Evision.Mat.t().

    Optional output derivative of tvec3 with regard to rvec1

  • dt3dt1: Evision.Mat.t().

    Optional output derivative of tvec3 with regard to tvec1

  • dt3dr2: Evision.Mat.t().

    Optional output derivative of tvec3 with regard to rvec2

  • dt3dt2: Evision.Mat.t().

    Optional output derivative of tvec3 with regard to tvec2

The functions compute: \f[\begin{array}{l} \texttt{rvec3} = \mathrm{rodrigues} ^{-1} \left ( \mathrm{rodrigues} ( \texttt{rvec2} ) \cdot \mathrm{rodrigues} ( \texttt{rvec1} ) \right ) \\ \texttt{tvec3} = \mathrm{rodrigues} ( \texttt{rvec2} ) \cdot \texttt{tvec1} + \texttt{tvec2} \end{array} ,\f] where \f$\mathrm{rodrigues}\f$ denotes a rotation vector to a rotation matrix transformation, and \f$\mathrm{rodrigues}^{-1}\f$ denotes the inverse transformation. See #Rodrigues for details. Also, the functions can compute the derivatives of the output vectors with regards to the input vectors (see #matMulDeriv ). The functions are used inside #stereoCalibrate but can also be used in your own code where Levenberg-Marquardt or another gradient-based solver is used to optimize a function that contains a matrix multiplication.

Python prototype (for reference only):

composeRT(rvec1, tvec1, rvec2, tvec2[, rvec3[, tvec3[, dr3dr1[, dr3dt1[, dr3dr2[, dr3dt2[, dt3dr1[, dt3dt1[, dt3dr2[, dt3dt2]]]]]]]]]]) -> rvec3, tvec3, dr3dr1, dr3dt1, dr3dr2, dr3dt2, dt3dr1, dt3dt1, dt3dr2, dt3dt2
Link to this function

composeRT(rvec1, tvec1, rvec2, tvec2, opts)

View Source

Combines two rotation-and-shift transformations.

Positional Arguments
  • rvec1: Evision.Mat.t().

    First rotation vector.

  • tvec1: Evision.Mat.t().

    First translation vector.

  • rvec2: Evision.Mat.t().

    Second rotation vector.

  • tvec2: Evision.Mat.t().

    Second translation vector.

Return
  • rvec3: Evision.Mat.t().

    Output rotation vector of the superposition.

  • tvec3: Evision.Mat.t().

    Output translation vector of the superposition.

  • dr3dr1: Evision.Mat.t().

    Optional output derivative of rvec3 with regard to rvec1

  • dr3dt1: Evision.Mat.t().

    Optional output derivative of rvec3 with regard to tvec1

  • dr3dr2: Evision.Mat.t().

    Optional output derivative of rvec3 with regard to rvec2

  • dr3dt2: Evision.Mat.t().

    Optional output derivative of rvec3 with regard to tvec2

  • dt3dr1: Evision.Mat.t().

    Optional output derivative of tvec3 with regard to rvec1

  • dt3dt1: Evision.Mat.t().

    Optional output derivative of tvec3 with regard to tvec1

  • dt3dr2: Evision.Mat.t().

    Optional output derivative of tvec3 with regard to rvec2

  • dt3dt2: Evision.Mat.t().

    Optional output derivative of tvec3 with regard to tvec2

The functions compute: \f[\begin{array}{l} \texttt{rvec3} = \mathrm{rodrigues} ^{-1} \left ( \mathrm{rodrigues} ( \texttt{rvec2} ) \cdot \mathrm{rodrigues} ( \texttt{rvec1} ) \right ) \\ \texttt{tvec3} = \mathrm{rodrigues} ( \texttt{rvec2} ) \cdot \texttt{tvec1} + \texttt{tvec2} \end{array} ,\f] where \f$\mathrm{rodrigues}\f$ denotes a rotation vector to a rotation matrix transformation, and \f$\mathrm{rodrigues}^{-1}\f$ denotes the inverse transformation. See #Rodrigues for details. Also, the functions can compute the derivatives of the output vectors with regards to the input vectors (see #matMulDeriv ). The functions are used inside #stereoCalibrate but can also be used in your own code where Levenberg-Marquardt or another gradient-based solver is used to optimize a function that contains a matrix multiplication.

Python prototype (for reference only):

composeRT(rvec1, tvec1, rvec2, tvec2[, rvec3[, tvec3[, dr3dr1[, dr3dt1[, dr3dr2[, dr3dt2[, dt3dr1[, dt3dt1[, dt3dr2[, dt3dt2]]]]]]]]]]) -> rvec3, tvec3, dr3dr1, dr3dt1, dr3dr2, dr3dt2, dt3dr1, dt3dt1, dt3dr2, dt3dt2
Link to this function

computeCorrespondEpilines(points, whichImage, f)

View Source
@spec computeCorrespondEpilines(
  Evision.Mat.maybe_mat_in(),
  integer(),
  Evision.Mat.maybe_mat_in()
) ::
  Evision.Mat.t() | {:error, String.t()}

For points in an image of a stereo pair, computes the corresponding epilines in the other image.

Positional Arguments
  • points: Evision.Mat.t().

    Input points. \f$N \times 1\f$ or \f$1 \times N\f$ matrix of type CV_32FC2 or vector\<Point2f> .

  • whichImage: int.

    Index of the image (1 or 2) that contains the points .

  • f: Evision.Mat.t().

    Fundamental matrix that can be estimated using #findFundamentalMat or #stereoRectify .

Return
  • lines: Evision.Mat.t().

    Output vector of the epipolar lines corresponding to the points in the other image. Each line \f$ax + by + c=0\f$ is encoded by 3 numbers \f$(a, b, c)\f$ .

For every point in one of the two images of a stereo pair, the function finds the equation of the corresponding epipolar line in the other image. From the fundamental matrix definition (see #findFundamentalMat ), line \f$l^{(2)}_i\f$ in the second image for the point \f$p^{(1)}_i\f$ in the first image (when whichImage=1 ) is computed as: \f[l^{(2)}_i = F p^{(1)}_i\f] And vice versa, when whichImage=2, \f$l^{(1)}_i\f$ is computed from \f$p^{(2)}_i\f$ as: \f[l^{(1)}_i = F^T p^{(2)}_i\f] Line coefficients are defined up to a scale. They are normalized so that \f$a_i^2+b_i^2=1\f$ .

Python prototype (for reference only):

computeCorrespondEpilines(points, whichImage, F[, lines]) -> lines
Link to this function

computeCorrespondEpilines(points, whichImage, f, opts)

View Source
@spec computeCorrespondEpilines(
  Evision.Mat.maybe_mat_in(),
  integer(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

For points in an image of a stereo pair, computes the corresponding epilines in the other image.

Positional Arguments
  • points: Evision.Mat.t().

    Input points. \f$N \times 1\f$ or \f$1 \times N\f$ matrix of type CV_32FC2 or vector\<Point2f> .

  • whichImage: int.

    Index of the image (1 or 2) that contains the points .

  • f: Evision.Mat.t().

    Fundamental matrix that can be estimated using #findFundamentalMat or #stereoRectify .

Return
  • lines: Evision.Mat.t().

    Output vector of the epipolar lines corresponding to the points in the other image. Each line \f$ax + by + c=0\f$ is encoded by 3 numbers \f$(a, b, c)\f$ .

For every point in one of the two images of a stereo pair, the function finds the equation of the corresponding epipolar line in the other image. From the fundamental matrix definition (see #findFundamentalMat ), line \f$l^{(2)}_i\f$ in the second image for the point \f$p^{(1)}_i\f$ in the first image (when whichImage=1 ) is computed as: \f[l^{(2)}_i = F p^{(1)}_i\f] And vice versa, when whichImage=2, \f$l^{(1)}_i\f$ is computed from \f$p^{(2)}_i\f$ as: \f[l^{(1)}_i = F^T p^{(2)}_i\f] Line coefficients are defined up to a scale. They are normalized so that \f$a_i^2+b_i^2=1\f$ .

Python prototype (for reference only):

computeCorrespondEpilines(points, whichImage, F[, lines]) -> lines
Link to this function

computeECC(templateImage, inputImage)

View Source
@spec computeECC(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in()) ::
  number() | {:error, String.t()}

Computes the Enhanced Correlation Coefficient value between two images @cite EP08 .

Positional Arguments
  • templateImage: Evision.Mat.t().

    single-channel template image; CV_8U or CV_32F array.

  • inputImage: Evision.Mat.t().

    single-channel input image to be warped to provide an image similar to templateImage, same type as templateImage.

Keyword Arguments
  • inputMask: Evision.Mat.t().

    An optional mask to indicate valid values of inputImage.

Return
  • retval: double

@sa findTransformECC

Python prototype (for reference only):

computeECC(templateImage, inputImage[, inputMask]) -> retval
Link to this function

computeECC(templateImage, inputImage, opts)

View Source
@spec computeECC(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  number() | {:error, String.t()}

Computes the Enhanced Correlation Coefficient value between two images @cite EP08 .

Positional Arguments
  • templateImage: Evision.Mat.t().

    single-channel template image; CV_8U or CV_32F array.

  • inputImage: Evision.Mat.t().

    single-channel input image to be warped to provide an image similar to templateImage, same type as templateImage.

Keyword Arguments
  • inputMask: Evision.Mat.t().

    An optional mask to indicate valid values of inputImage.

Return
  • retval: double

@sa findTransformECC

Python prototype (for reference only):

computeECC(templateImage, inputImage[, inputMask]) -> retval
Link to this function

connectedComponents(image)

View Source
@spec connectedComponents(Evision.Mat.maybe_mat_in()) ::
  {integer(), Evision.Mat.t()} | {:error, String.t()}

connectedComponents

Positional Arguments
  • image: Evision.Mat.t().

    the 8-bit single-channel image to be labeled

Keyword Arguments
  • connectivity: int.

    8 or 4 for 8-way or 4-way connectivity respectively

  • ltype: int.

    output image label type. Currently CV_32S and CV_16U are supported.

Return
  • retval: int

  • labels: Evision.Mat.t().

    destination labeled image

Has overloading in C++

Python prototype (for reference only):

connectedComponents(image[, labels[, connectivity[, ltype]]]) -> retval, labels
Link to this function

connectedComponents(image, opts)

View Source
@spec connectedComponents(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  {integer(), Evision.Mat.t()} | {:error, String.t()}

connectedComponents

Positional Arguments
  • image: Evision.Mat.t().

    the 8-bit single-channel image to be labeled

Keyword Arguments
  • connectivity: int.

    8 or 4 for 8-way or 4-way connectivity respectively

  • ltype: int.

    output image label type. Currently CV_32S and CV_16U are supported.

Return
  • retval: int

  • labels: Evision.Mat.t().

    destination labeled image

Has overloading in C++

Python prototype (for reference only):

connectedComponents(image[, labels[, connectivity[, ltype]]]) -> retval, labels
Link to this function

connectedComponentsWithAlgorithm(image, connectivity, ltype, ccltype)

View Source
@spec connectedComponentsWithAlgorithm(
  Evision.Mat.maybe_mat_in(),
  integer(),
  integer(),
  integer()
) ::
  {integer(), Evision.Mat.t()} | {:error, String.t()}

computes the connected components labeled image of boolean image

Positional Arguments
  • image: Evision.Mat.t().

    the 8-bit single-channel image to be labeled

  • connectivity: int.

    8 or 4 for 8-way or 4-way connectivity respectively

  • ltype: int.

    output image label type. Currently CV_32S and CV_16U are supported.

  • ccltype: int.

    connected components algorithm type (see the #ConnectedComponentsAlgorithmsTypes).

Return
  • retval: int

  • labels: Evision.Mat.t().

    destination labeled image

image with 4 or 8 way connectivity - returns N, the total number of labels [0, N-1] where 0 represents the background label. ltype specifies the output label image type, an important consideration based on the total number of labels or alternatively the total number of pixels in the source image. ccltype specifies the connected components labeling algorithm to use, currently Bolelli (Spaghetti) @cite Bolelli2019, Grana (BBDT) @cite Grana2010 and Wu's (SAUF) @cite Wu2009 algorithms are supported, see the #ConnectedComponentsAlgorithmsTypes for details. Note that SAUF algorithm forces a row major ordering of labels while Spaghetti and BBDT do not. This function uses parallel version of the algorithms if at least one allowed parallel framework is enabled and if the rows of the image are at least twice the number returned by #getNumberOfCPUs.

Python prototype (for reference only):

connectedComponentsWithAlgorithm(image, connectivity, ltype, ccltype[, labels]) -> retval, labels
Link to this function

connectedComponentsWithAlgorithm(image, connectivity, ltype, ccltype, opts)

View Source
@spec connectedComponentsWithAlgorithm(
  Evision.Mat.maybe_mat_in(),
  integer(),
  integer(),
  integer(),
  [{atom(), term()}, ...] | nil
) :: {integer(), Evision.Mat.t()} | {:error, String.t()}

computes the connected components labeled image of boolean image

Positional Arguments
  • image: Evision.Mat.t().

    the 8-bit single-channel image to be labeled

  • connectivity: int.

    8 or 4 for 8-way or 4-way connectivity respectively

  • ltype: int.

    output image label type. Currently CV_32S and CV_16U are supported.

  • ccltype: int.

    connected components algorithm type (see the #ConnectedComponentsAlgorithmsTypes).

Return
  • retval: int

  • labels: Evision.Mat.t().

    destination labeled image

image with 4 or 8 way connectivity - returns N, the total number of labels [0, N-1] where 0 represents the background label. ltype specifies the output label image type, an important consideration based on the total number of labels or alternatively the total number of pixels in the source image. ccltype specifies the connected components labeling algorithm to use, currently Bolelli (Spaghetti) @cite Bolelli2019, Grana (BBDT) @cite Grana2010 and Wu's (SAUF) @cite Wu2009 algorithms are supported, see the #ConnectedComponentsAlgorithmsTypes for details. Note that SAUF algorithm forces a row major ordering of labels while Spaghetti and BBDT do not. This function uses parallel version of the algorithms if at least one allowed parallel framework is enabled and if the rows of the image are at least twice the number returned by #getNumberOfCPUs.

Python prototype (for reference only):

connectedComponentsWithAlgorithm(image, connectivity, ltype, ccltype[, labels]) -> retval, labels
Link to this function

connectedComponentsWithStats(image)

View Source
@spec connectedComponentsWithStats(Evision.Mat.maybe_mat_in()) ::
  {integer(), Evision.Mat.t(), Evision.Mat.t(), Evision.Mat.t()}
  | {:error, String.t()}

connectedComponentsWithStats

Positional Arguments
  • image: Evision.Mat.t().

    the 8-bit single-channel image to be labeled

Keyword Arguments
  • connectivity: int.

    8 or 4 for 8-way or 4-way connectivity respectively

  • ltype: int.

    output image label type. Currently CV_32S and CV_16U are supported.

Return
  • retval: int

  • labels: Evision.Mat.t().

    destination labeled image

  • stats: Evision.Mat.t().

    statistics output for each label, including the background label. Statistics are accessed via stats(label, COLUMN) where COLUMN is one of #ConnectedComponentsTypes, selecting the statistic. The data type is CV_32S.

  • centroids: Evision.Mat.t().

    centroid output for each label, including the background label. Centroids are accessed via centroids(label, 0) for x and centroids(label, 1) for y. The data type CV_64F.

Has overloading in C++

Python prototype (for reference only):

connectedComponentsWithStats(image[, labels[, stats[, centroids[, connectivity[, ltype]]]]]) -> retval, labels, stats, centroids
Link to this function

connectedComponentsWithStats(image, opts)

View Source
@spec connectedComponentsWithStats(
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  {integer(), Evision.Mat.t(), Evision.Mat.t(), Evision.Mat.t()}
  | {:error, String.t()}

connectedComponentsWithStats

Positional Arguments
  • image: Evision.Mat.t().

    the 8-bit single-channel image to be labeled

Keyword Arguments
  • connectivity: int.

    8 or 4 for 8-way or 4-way connectivity respectively

  • ltype: int.

    output image label type. Currently CV_32S and CV_16U are supported.

Return
  • retval: int

  • labels: Evision.Mat.t().

    destination labeled image

  • stats: Evision.Mat.t().

    statistics output for each label, including the background label. Statistics are accessed via stats(label, COLUMN) where COLUMN is one of #ConnectedComponentsTypes, selecting the statistic. The data type is CV_32S.

  • centroids: Evision.Mat.t().

    centroid output for each label, including the background label. Centroids are accessed via centroids(label, 0) for x and centroids(label, 1) for y. The data type CV_64F.

Has overloading in C++

Python prototype (for reference only):

connectedComponentsWithStats(image[, labels[, stats[, centroids[, connectivity[, ltype]]]]]) -> retval, labels, stats, centroids
Link to this function

connectedComponentsWithStatsWithAlgorithm(image, connectivity, ltype, ccltype)

View Source
@spec connectedComponentsWithStatsWithAlgorithm(
  Evision.Mat.maybe_mat_in(),
  integer(),
  integer(),
  integer()
) ::
  {integer(), Evision.Mat.t(), Evision.Mat.t(), Evision.Mat.t()}
  | {:error, String.t()}

computes the connected components labeled image of boolean image and also produces a statistics output for each label

Positional Arguments
  • image: Evision.Mat.t().

    the 8-bit single-channel image to be labeled

  • connectivity: int.

    8 or 4 for 8-way or 4-way connectivity respectively

  • ltype: int.

    output image label type. Currently CV_32S and CV_16U are supported.

  • ccltype: int.

    connected components algorithm type (see #ConnectedComponentsAlgorithmsTypes).

Return
  • retval: int

  • labels: Evision.Mat.t().

    destination labeled image

  • stats: Evision.Mat.t().

    statistics output for each label, including the background label. Statistics are accessed via stats(label, COLUMN) where COLUMN is one of #ConnectedComponentsTypes, selecting the statistic. The data type is CV_32S.

  • centroids: Evision.Mat.t().

    centroid output for each label, including the background label. Centroids are accessed via centroids(label, 0) for x and centroids(label, 1) for y. The data type CV_64F.

image with 4 or 8 way connectivity - returns N, the total number of labels [0, N-1] where 0 represents the background label. ltype specifies the output label image type, an important consideration based on the total number of labels or alternatively the total number of pixels in the source image. ccltype specifies the connected components labeling algorithm to use, currently Bolelli (Spaghetti) @cite Bolelli2019, Grana (BBDT) @cite Grana2010 and Wu's (SAUF) @cite Wu2009 algorithms are supported, see the #ConnectedComponentsAlgorithmsTypes for details. Note that SAUF algorithm forces a row major ordering of labels while Spaghetti and BBDT do not. This function uses parallel version of the algorithms (statistics included) if at least one allowed parallel framework is enabled and if the rows of the image are at least twice the number returned by #getNumberOfCPUs.

Python prototype (for reference only):

connectedComponentsWithStatsWithAlgorithm(image, connectivity, ltype, ccltype[, labels[, stats[, centroids]]]) -> retval, labels, stats, centroids
Link to this function

connectedComponentsWithStatsWithAlgorithm(image, connectivity, ltype, ccltype, opts)

View Source
@spec connectedComponentsWithStatsWithAlgorithm(
  Evision.Mat.maybe_mat_in(),
  integer(),
  integer(),
  integer(),
  [{atom(), term()}, ...] | nil
) ::
  {integer(), Evision.Mat.t(), Evision.Mat.t(), Evision.Mat.t()}
  | {:error, String.t()}

computes the connected components labeled image of boolean image and also produces a statistics output for each label

Positional Arguments
  • image: Evision.Mat.t().

    the 8-bit single-channel image to be labeled

  • connectivity: int.

    8 or 4 for 8-way or 4-way connectivity respectively

  • ltype: int.

    output image label type. Currently CV_32S and CV_16U are supported.

  • ccltype: int.

    connected components algorithm type (see #ConnectedComponentsAlgorithmsTypes).

Return
  • retval: int

  • labels: Evision.Mat.t().

    destination labeled image

  • stats: Evision.Mat.t().

    statistics output for each label, including the background label. Statistics are accessed via stats(label, COLUMN) where COLUMN is one of #ConnectedComponentsTypes, selecting the statistic. The data type is CV_32S.

  • centroids: Evision.Mat.t().

    centroid output for each label, including the background label. Centroids are accessed via centroids(label, 0) for x and centroids(label, 1) for y. The data type CV_64F.

image with 4 or 8 way connectivity - returns N, the total number of labels [0, N-1] where 0 represents the background label. ltype specifies the output label image type, an important consideration based on the total number of labels or alternatively the total number of pixels in the source image. ccltype specifies the connected components labeling algorithm to use, currently Bolelli (Spaghetti) @cite Bolelli2019, Grana (BBDT) @cite Grana2010 and Wu's (SAUF) @cite Wu2009 algorithms are supported, see the #ConnectedComponentsAlgorithmsTypes for details. Note that SAUF algorithm forces a row major ordering of labels while Spaghetti and BBDT do not. This function uses parallel version of the algorithms (statistics included) if at least one allowed parallel framework is enabled and if the rows of the image are at least twice the number returned by #getNumberOfCPUs.

Python prototype (for reference only):

connectedComponentsWithStatsWithAlgorithm(image, connectivity, ltype, ccltype[, labels[, stats[, centroids]]]) -> retval, labels, stats, centroids
@spec contourArea(Evision.Mat.maybe_mat_in()) :: number() | {:error, String.t()}

Calculates a contour area.

Positional Arguments
  • contour: Evision.Mat.t().

    Input vector of 2D points (contour vertices), stored in std::vector or Mat.

Keyword Arguments
  • oriented: bool.

    Oriented area flag. If it is true, the function returns a signed area value, depending on the contour orientation (clockwise or counter-clockwise). Using this feature you can determine orientation of a contour by taking the sign of an area. By default, the parameter is false, which means that the absolute value is returned.

Return
  • retval: double

The function computes a contour area. Similarly to moments , the area is computed using the Green formula. Thus, the returned area and the number of non-zero pixels, if you draw the contour using #drawContours or #fillPoly , can be different. Also, the function will most certainly give a wrong results for contours with self-intersections. Example:

vector<Point> contour;
contour.push_back(Point2f(0, 0));
contour.push_back(Point2f(10, 0));
contour.push_back(Point2f(10, 10));
contour.push_back(Point2f(5, 4));
double area0 = contourArea(contour);
vector<Point> approx;
approxPolyDP(contour, approx, 5, true);
double area1 = contourArea(approx);
cout << "area0 =" << area0 << endl <<
"area1 =" << area1 << endl <<
"approx poly vertices" << approx.size() << endl;

Python prototype (for reference only):

contourArea(contour[, oriented]) -> retval
Link to this function

contourArea(contour, opts)

View Source
@spec contourArea(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  number() | {:error, String.t()}

Calculates a contour area.

Positional Arguments
  • contour: Evision.Mat.t().

    Input vector of 2D points (contour vertices), stored in std::vector or Mat.

Keyword Arguments
  • oriented: bool.

    Oriented area flag. If it is true, the function returns a signed area value, depending on the contour orientation (clockwise or counter-clockwise). Using this feature you can determine orientation of a contour by taking the sign of an area. By default, the parameter is false, which means that the absolute value is returned.

Return
  • retval: double

The function computes a contour area. Similarly to moments , the area is computed using the Green formula. Thus, the returned area and the number of non-zero pixels, if you draw the contour using #drawContours or #fillPoly , can be different. Also, the function will most certainly give a wrong results for contours with self-intersections. Example:

vector<Point> contour;
contour.push_back(Point2f(0, 0));
contour.push_back(Point2f(10, 0));
contour.push_back(Point2f(10, 10));
contour.push_back(Point2f(5, 4));
double area0 = contourArea(contour);
vector<Point> approx;
approxPolyDP(contour, approx, 5, true);
double area1 = contourArea(approx);
cout << "area0 =" << area0 << endl <<
"area1 =" << area1 << endl <<
"approx poly vertices" << approx.size() << endl;

Python prototype (for reference only):

contourArea(contour[, oriented]) -> retval
@spec convertFp16(Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

Converts an array to half precision floating number.

Positional Arguments
  • src: Evision.Mat.t().

    input array.

Return
  • dst: Evision.Mat.t().

    output array.

This function converts FP32 (single precision floating point) from/to FP16 (half precision floating point). CV_16S format is used to represent FP16 data. There are two use modes (src -> dst): CV_32F -> CV_16S and CV_16S -> CV_32F. The input array has to have type of CV_32F or CV_16S to represent the bit depth. If the input array is neither of them, the function will raise an error. The format of half precision floating point is defined in IEEE 754-2008.

Python prototype (for reference only):

convertFp16(src[, dst]) -> dst
@spec convertFp16(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Converts an array to half precision floating number.

Positional Arguments
  • src: Evision.Mat.t().

    input array.

Return
  • dst: Evision.Mat.t().

    output array.

This function converts FP32 (single precision floating point) from/to FP16 (half precision floating point). CV_16S format is used to represent FP16 data. There are two use modes (src -> dst): CV_32F -> CV_16S and CV_16S -> CV_32F. The input array has to have type of CV_32F or CV_16S to represent the bit depth. If the input array is neither of them, the function will raise an error. The format of half precision floating point is defined in IEEE 754-2008.

Python prototype (for reference only):

convertFp16(src[, dst]) -> dst
Link to this function

convertMaps(map1, map2, dstmap1type)

View Source

Converts image transformation maps from one representation to another.

Positional Arguments
  • map1: Evision.Mat.t().

    The first input map of type CV_16SC2, CV_32FC1, or CV_32FC2 .

  • map2: Evision.Mat.t().

    The second input map of type CV_16UC1, CV_32FC1, or none (empty matrix), respectively.

  • dstmap1type: int.

    Type of the first output map that should be CV_16SC2, CV_32FC1, or CV_32FC2 .

Keyword Arguments
  • nninterpolation: bool.

    Flag indicating whether the fixed-point maps are used for the nearest-neighbor or for a more complex interpolation.

Return
  • dstmap1: Evision.Mat.t().

    The first output map that has the type dstmap1type and the same size as src .

  • dstmap2: Evision.Mat.t().

    The second output map.

The function converts a pair of maps for remap from one representation to another. The following options ( (map1.type(), map2.type()) \f$\rightarrow\f$ (dstmap1.type(), dstmap2.type()) ) are supported:

  • \f$\texttt{(CV_32FC1, CV_32FC1)} \rightarrow \texttt{(CV_16SC2, CV_16UC1)}\f$. This is the most frequently used conversion operation, in which the original floating-point maps (see #remap) are converted to a more compact and much faster fixed-point representation. The first output array contains the rounded coordinates and the second array (created only when nninterpolation=false ) contains indices in the interpolation tables.

  • \f$\texttt{(CV_32FC2)} \rightarrow \texttt{(CV_16SC2, CV_16UC1)}\f$. The same as above but the original maps are stored in one 2-channel matrix.

  • Reverse conversion. Obviously, the reconstructed floating-point maps will not be exactly the same as the originals.

@sa remap, undistort, initUndistortRectifyMap

Python prototype (for reference only):

convertMaps(map1, map2, dstmap1type[, dstmap1[, dstmap2[, nninterpolation]]]) -> dstmap1, dstmap2
Link to this function

convertMaps(map1, map2, dstmap1type, opts)

View Source
@spec convertMaps(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  integer(),
  [{atom(), term()}, ...] | nil
) :: {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Converts image transformation maps from one representation to another.

Positional Arguments
  • map1: Evision.Mat.t().

    The first input map of type CV_16SC2, CV_32FC1, or CV_32FC2 .

  • map2: Evision.Mat.t().

    The second input map of type CV_16UC1, CV_32FC1, or none (empty matrix), respectively.

  • dstmap1type: int.

    Type of the first output map that should be CV_16SC2, CV_32FC1, or CV_32FC2 .

Keyword Arguments
  • nninterpolation: bool.

    Flag indicating whether the fixed-point maps are used for the nearest-neighbor or for a more complex interpolation.

Return
  • dstmap1: Evision.Mat.t().

    The first output map that has the type dstmap1type and the same size as src .

  • dstmap2: Evision.Mat.t().

    The second output map.

The function converts a pair of maps for remap from one representation to another. The following options ( (map1.type(), map2.type()) \f$\rightarrow\f$ (dstmap1.type(), dstmap2.type()) ) are supported:

  • \f$\texttt{(CV_32FC1, CV_32FC1)} \rightarrow \texttt{(CV_16SC2, CV_16UC1)}\f$. This is the most frequently used conversion operation, in which the original floating-point maps (see #remap) are converted to a more compact and much faster fixed-point representation. The first output array contains the rounded coordinates and the second array (created only when nninterpolation=false ) contains indices in the interpolation tables.

  • \f$\texttt{(CV_32FC2)} \rightarrow \texttt{(CV_16SC2, CV_16UC1)}\f$. The same as above but the original maps are stored in one 2-channel matrix.

  • Reverse conversion. Obviously, the reconstructed floating-point maps will not be exactly the same as the originals.

@sa remap, undistort, initUndistortRectifyMap

Python prototype (for reference only):

convertMaps(map1, map2, dstmap1type[, dstmap1[, dstmap2[, nninterpolation]]]) -> dstmap1, dstmap2
Link to this function

convertPointsFromHomogeneous(src)

View Source
@spec convertPointsFromHomogeneous(Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

Converts points from homogeneous to Euclidean space.

Positional Arguments
  • src: Evision.Mat.t().

    Input vector of N-dimensional points.

Return
  • dst: Evision.Mat.t().

    Output vector of N-1-dimensional points.

The function converts points homogeneous to Euclidean space using perspective projection. That is, each point (x1, x2, ... x(n-1), xn) is converted to (x1/xn, x2/xn, ..., x(n-1)/xn). When xn=0, the output point coordinates will be (0,0,0,...).

Python prototype (for reference only):

convertPointsFromHomogeneous(src[, dst]) -> dst
Link to this function

convertPointsFromHomogeneous(src, opts)

View Source
@spec convertPointsFromHomogeneous(
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Converts points from homogeneous to Euclidean space.

Positional Arguments
  • src: Evision.Mat.t().

    Input vector of N-dimensional points.

Return
  • dst: Evision.Mat.t().

    Output vector of N-1-dimensional points.

The function converts points homogeneous to Euclidean space using perspective projection. That is, each point (x1, x2, ... x(n-1), xn) is converted to (x1/xn, x2/xn, ..., x(n-1)/xn). When xn=0, the output point coordinates will be (0,0,0,...).

Python prototype (for reference only):

convertPointsFromHomogeneous(src[, dst]) -> dst
Link to this function

convertPointsToHomogeneous(src)

View Source
@spec convertPointsToHomogeneous(Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

Converts points from Euclidean to homogeneous space.

Positional Arguments
  • src: Evision.Mat.t().

    Input vector of N-dimensional points.

Return
  • dst: Evision.Mat.t().

    Output vector of N+1-dimensional points.

The function converts points from Euclidean to homogeneous space by appending 1's to the tuple of point coordinates. That is, each point (x1, x2, ..., xn) is converted to (x1, x2, ..., xn, 1).

Python prototype (for reference only):

convertPointsToHomogeneous(src[, dst]) -> dst
Link to this function

convertPointsToHomogeneous(src, opts)

View Source
@spec convertPointsToHomogeneous(
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Converts points from Euclidean to homogeneous space.

Positional Arguments
  • src: Evision.Mat.t().

    Input vector of N-dimensional points.

Return
  • dst: Evision.Mat.t().

    Output vector of N+1-dimensional points.

The function converts points from Euclidean to homogeneous space by appending 1's to the tuple of point coordinates. That is, each point (x1, x2, ..., xn) is converted to (x1, x2, ..., xn, 1).

Python prototype (for reference only):

convertPointsToHomogeneous(src[, dst]) -> dst
@spec convertScaleAbs(Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

Scales, calculates absolute values, and converts the result to 8-bit.

Positional Arguments
  • src: Evision.Mat.t().

    input array.

Keyword Arguments
  • alpha: double.

    optional scale factor.

  • beta: double.

    optional delta added to the scaled values.

Return
  • dst: Evision.Mat.t().

    output array.

On each element of the input array, the function convertScaleAbs performs three operations sequentially: scaling, taking an absolute value, conversion to an unsigned 8-bit type: \f[\texttt{dst} (I)= \texttt{saturate\_cast<uchar>} (| \texttt{src} (I)* \texttt{alpha} + \texttt{beta} |)\f] In case of multi-channel arrays, the function processes each channel independently. When the output is not 8-bit, the operation can be emulated by calling the Mat::convertTo method (or by using matrix expressions) and then by calculating an absolute value of the result. For example:

Mat_<float> A(30,30);
randu(A, Scalar(-100), Scalar(100));
Mat_<float> B = A*5 + 3;
B = abs(B);
// Mat_<float> B = abs(A*5+3) will also do the job,
// but it will allocate a temporary matrix

@sa Mat::convertTo, cv::abs(const Mat&)

Python prototype (for reference only):

convertScaleAbs(src[, dst[, alpha[, beta]]]) -> dst
Link to this function

convertScaleAbs(src, opts)

View Source
@spec convertScaleAbs(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Scales, calculates absolute values, and converts the result to 8-bit.

Positional Arguments
  • src: Evision.Mat.t().

    input array.

Keyword Arguments
  • alpha: double.

    optional scale factor.

  • beta: double.

    optional delta added to the scaled values.

Return
  • dst: Evision.Mat.t().

    output array.

On each element of the input array, the function convertScaleAbs performs three operations sequentially: scaling, taking an absolute value, conversion to an unsigned 8-bit type: \f[\texttt{dst} (I)= \texttt{saturate\_cast<uchar>} (| \texttt{src} (I)* \texttt{alpha} + \texttt{beta} |)\f] In case of multi-channel arrays, the function processes each channel independently. When the output is not 8-bit, the operation can be emulated by calling the Mat::convertTo method (or by using matrix expressions) and then by calculating an absolute value of the result. For example:

Mat_<float> A(30,30);
randu(A, Scalar(-100), Scalar(100));
Mat_<float> B = A*5 + 3;
B = abs(B);
// Mat_<float> B = abs(A*5+3) will also do the job,
// but it will allocate a temporary matrix

@sa Mat::convertTo, cv::abs(const Mat&)

Python prototype (for reference only):

convertScaleAbs(src[, dst[, alpha[, beta]]]) -> dst
@spec convexHull(Evision.Mat.maybe_mat_in()) :: Evision.Mat.t() | {:error, String.t()}

Finds the convex hull of a point set.

Positional Arguments
  • points: Evision.Mat.t().

    Input 2D point set, stored in std::vector or Mat.

Keyword Arguments
  • clockwise: bool.

    Orientation flag. If it is true, the output convex hull is oriented clockwise. Otherwise, it is oriented counter-clockwise. The assumed coordinate system has its X axis pointing to the right, and its Y axis pointing upwards.

  • returnPoints: bool.

    Operation flag. In case of a matrix, when the flag is true, the function returns convex hull points. Otherwise, it returns indices of the convex hull points. When the output array is std::vector, the flag is ignored, and the output depends on the type of the vector: std::vector\<int> implies returnPoints=false, std::vector\<Point> implies returnPoints=true.

Return
  • hull: Evision.Mat.t().

    Output convex hull. It is either an integer vector of indices or vector of points. In the first case, the hull elements are 0-based indices of the convex hull points in the original array (since the set of convex hull points is a subset of the original point set). In the second case, hull elements are the convex hull points themselves.

The function cv::convexHull finds the convex hull of a 2D point set using the Sklansky's algorithm @cite Sklansky82 that has O(N logN) complexity in the current implementation.

Note: points and hull should be different arrays, inplace processing isn't supported. Check @ref tutorial_hull "the corresponding tutorial" for more details. useful links: https://www.learnopencv.com/convex-hull-using-opencv-in-python-and-c/

Python prototype (for reference only):

convexHull(points[, hull[, clockwise[, returnPoints]]]) -> hull
Link to this function

convexHull(points, opts)

View Source
@spec convexHull(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Finds the convex hull of a point set.

Positional Arguments
  • points: Evision.Mat.t().

    Input 2D point set, stored in std::vector or Mat.

Keyword Arguments
  • clockwise: bool.

    Orientation flag. If it is true, the output convex hull is oriented clockwise. Otherwise, it is oriented counter-clockwise. The assumed coordinate system has its X axis pointing to the right, and its Y axis pointing upwards.

  • returnPoints: bool.

    Operation flag. In case of a matrix, when the flag is true, the function returns convex hull points. Otherwise, it returns indices of the convex hull points. When the output array is std::vector, the flag is ignored, and the output depends on the type of the vector: std::vector\<int> implies returnPoints=false, std::vector\<Point> implies returnPoints=true.

Return
  • hull: Evision.Mat.t().

    Output convex hull. It is either an integer vector of indices or vector of points. In the first case, the hull elements are 0-based indices of the convex hull points in the original array (since the set of convex hull points is a subset of the original point set). In the second case, hull elements are the convex hull points themselves.

The function cv::convexHull finds the convex hull of a 2D point set using the Sklansky's algorithm @cite Sklansky82 that has O(N logN) complexity in the current implementation.

Note: points and hull should be different arrays, inplace processing isn't supported. Check @ref tutorial_hull "the corresponding tutorial" for more details. useful links: https://www.learnopencv.com/convex-hull-using-opencv-in-python-and-c/

Python prototype (for reference only):

convexHull(points[, hull[, clockwise[, returnPoints]]]) -> hull
Link to this function

convexityDefects(contour, convexhull)

View Source
@spec convexityDefects(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

Finds the convexity defects of a contour.

Positional Arguments
  • contour: Evision.Mat.t().

    Input contour.

  • convexhull: Evision.Mat.t().

    Convex hull obtained using convexHull that should contain indices of the contour points that make the hull.

Return
  • convexityDefects: Evision.Mat.t().

    The output vector of convexity defects. In C++ and the new Python/Java interface each convexity defect is represented as 4-element integer vector (a.k.a. #Vec4i): (start_index, end_index, farthest_pt_index, fixpt_depth), where indices are 0-based indices in the original contour of the convexity defect beginning, end and the farthest point, and fixpt_depth is fixed-point approximation (with 8 fractional bits) of the distance between the farthest contour point and the hull. That is, to get the floating-point value of the depth will be fixpt_depth/256.0.

The figure below displays convexity defects of a hand contour: image

Python prototype (for reference only):

convexityDefects(contour, convexhull[, convexityDefects]) -> convexityDefects
Link to this function

convexityDefects(contour, convexhull, opts)

View Source
@spec convexityDefects(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Finds the convexity defects of a contour.

Positional Arguments
  • contour: Evision.Mat.t().

    Input contour.

  • convexhull: Evision.Mat.t().

    Convex hull obtained using convexHull that should contain indices of the contour points that make the hull.

Return
  • convexityDefects: Evision.Mat.t().

    The output vector of convexity defects. In C++ and the new Python/Java interface each convexity defect is represented as 4-element integer vector (a.k.a. #Vec4i): (start_index, end_index, farthest_pt_index, fixpt_depth), where indices are 0-based indices in the original contour of the convexity defect beginning, end and the farthest point, and fixpt_depth is fixed-point approximation (with 8 fractional bits) of the distance between the farthest contour point and the hull. That is, to get the floating-point value of the depth will be fixpt_depth/256.0.

The figure below displays convexity defects of a hand contour: image

Python prototype (for reference only):

convexityDefects(contour, convexhull[, convexityDefects]) -> convexityDefects
Link to this function

copyMakeBorder(src, top, bottom, left, right, borderType)

View Source
@spec copyMakeBorder(
  Evision.Mat.maybe_mat_in(),
  integer(),
  integer(),
  integer(),
  integer(),
  integer()
) ::
  Evision.Mat.t() | {:error, String.t()}

Forms a border around an image.

Positional Arguments
  • src: Evision.Mat.t().

    Source image.

  • top: int.

    the top pixels

  • bottom: int.

    the bottom pixels

  • left: int.

    the left pixels

  • right: int.

    Parameter specifying how many pixels in each direction from the source image rectangle to extrapolate. For example, top=1, bottom=1, left=1, right=1 mean that 1 pixel-wide border needs to be built.

  • borderType: int.

    Border type. See borderInterpolate for details.

Keyword Arguments
  • value: Scalar.

    Border value if borderType==BORDER_CONSTANT .

Return
  • dst: Evision.Mat.t().

    Destination image of the same type as src and the size Size(src.cols+left+right, src.rows+top+bottom) .

The function copies the source image into the middle of the destination image. The areas to the left, to the right, above and below the copied source image will be filled with extrapolated pixels. This is not what filtering functions based on it do (they extrapolate pixels on-fly), but what other more complex functions, including your own, may do to simplify image boundary handling. The function supports the mode when src is already in the middle of dst . In this case, the function does not copy src itself but simply constructs the border, for example:

// let border be the same in all directions
int border=2;
// constructs a larger image to fit both the image and the border
Mat gray_buf(rgb.rows + border*2, rgb.cols + border*2, rgb.depth());
// select the middle part of it w/o copying data
Mat gray(gray_canvas, Rect(border, border, rgb.cols, rgb.rows));
// convert image from RGB to grayscale
cvtColor(rgb, gray, COLOR_RGB2GRAY);
// form a border in-place
copyMakeBorder(gray, gray_buf, border, border,
border, border, BORDER_REPLICATE);
// now do some custom filtering ...
...

Note: When the source image is a part (ROI) of a bigger image, the function will try to use the pixels outside of the ROI to form a border. To disable this feature and always do extrapolation, as if src was not a ROI, use borderType | #BORDER_ISOLATED.

@sa borderInterpolate

Python prototype (for reference only):

copyMakeBorder(src, top, bottom, left, right, borderType[, dst[, value]]) -> dst
Link to this function

copyMakeBorder(src, top, bottom, left, right, borderType, opts)

View Source
@spec copyMakeBorder(
  Evision.Mat.maybe_mat_in(),
  integer(),
  integer(),
  integer(),
  integer(),
  integer(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Forms a border around an image.

Positional Arguments
  • src: Evision.Mat.t().

    Source image.

  • top: int.

    the top pixels

  • bottom: int.

    the bottom pixels

  • left: int.

    the left pixels

  • right: int.

    Parameter specifying how many pixels in each direction from the source image rectangle to extrapolate. For example, top=1, bottom=1, left=1, right=1 mean that 1 pixel-wide border needs to be built.

  • borderType: int.

    Border type. See borderInterpolate for details.

Keyword Arguments
  • value: Scalar.

    Border value if borderType==BORDER_CONSTANT .

Return
  • dst: Evision.Mat.t().

    Destination image of the same type as src and the size Size(src.cols+left+right, src.rows+top+bottom) .

The function copies the source image into the middle of the destination image. The areas to the left, to the right, above and below the copied source image will be filled with extrapolated pixels. This is not what filtering functions based on it do (they extrapolate pixels on-fly), but what other more complex functions, including your own, may do to simplify image boundary handling. The function supports the mode when src is already in the middle of dst . In this case, the function does not copy src itself but simply constructs the border, for example:

// let border be the same in all directions
int border=2;
// constructs a larger image to fit both the image and the border
Mat gray_buf(rgb.rows + border*2, rgb.cols + border*2, rgb.depth());
// select the middle part of it w/o copying data
Mat gray(gray_canvas, Rect(border, border, rgb.cols, rgb.rows));
// convert image from RGB to grayscale
cvtColor(rgb, gray, COLOR_RGB2GRAY);
// form a border in-place
copyMakeBorder(gray, gray_buf, border, border,
border, border, BORDER_REPLICATE);
// now do some custom filtering ...
...

Note: When the source image is a part (ROI) of a bigger image, the function will try to use the pixels outside of the ROI to form a border. To disable this feature and always do extrapolation, as if src was not a ROI, use borderType | #BORDER_ISOLATED.

@sa borderInterpolate

Python prototype (for reference only):

copyMakeBorder(src, top, bottom, left, right, borderType[, dst[, value]]) -> dst

This is an overloaded member function, provided for convenience (python) Copies the matrix to another one. When the operation mask is specified, if the Mat::create call shown above reallocates the matrix, the newly allocated matrix is initialized with all zeros before copying the data.

Positional Arguments
  • src: Evision.Mat.t().

    source matrix.

  • mask: Evision.Mat.t().

    Operation mask of the same size as *this. Its non-zero elements indicate which matrix elements need to be copied. The mask has to be of type CV_8U and can have 1 or multiple channels.

Return
  • dst: Evision.Mat.t().

    Destination matrix. If it does not have a proper size or type before the operation, it is reallocated.

Python prototype (for reference only):

copyTo(src, mask[, dst]) -> dst
@spec copyTo(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

This is an overloaded member function, provided for convenience (python) Copies the matrix to another one. When the operation mask is specified, if the Mat::create call shown above reallocates the matrix, the newly allocated matrix is initialized with all zeros before copying the data.

Positional Arguments
  • src: Evision.Mat.t().

    source matrix.

  • mask: Evision.Mat.t().

    Operation mask of the same size as *this. Its non-zero elements indicate which matrix elements need to be copied. The mask has to be of type CV_8U and can have 1 or multiple channels.

Return
  • dst: Evision.Mat.t().

    Destination matrix. If it does not have a proper size or type before the operation, it is reallocated.

Python prototype (for reference only):

copyTo(src, mask[, dst]) -> dst
Link to this function

cornerEigenValsAndVecs(src, blockSize, ksize)

View Source
@spec cornerEigenValsAndVecs(Evision.Mat.maybe_mat_in(), integer(), integer()) ::
  Evision.Mat.t() | {:error, String.t()}

Calculates eigenvalues and eigenvectors of image blocks for corner detection.

Positional Arguments
  • src: Evision.Mat.t().

    Input single-channel 8-bit or floating-point image.

  • blockSize: int.

    Neighborhood size (see details below).

  • ksize: int.

    Aperture parameter for the Sobel operator.

Keyword Arguments
  • borderType: int.

    Pixel extrapolation method. See #BorderTypes. #BORDER_WRAP is not supported.

Return
  • dst: Evision.Mat.t().

    Image to store the results. It has the same size as src and the type CV_32FC(6) .

For every pixel \f$p\f$ , the function cornerEigenValsAndVecs considers a blockSize \f$\times\f$ blockSize neighborhood \f$S(p)\f$ . It calculates the covariation matrix of derivatives over the neighborhood as: \f[M = \begin{bmatrix} \sum _{S(p)}(dI/dx)^2 & \sum _{S(p)}dI/dx dI/dy \\ \sum _{S(p)}dI/dx dI/dy & \sum _{S(p)}(dI/dy)^2 \end{bmatrix}\f] where the derivatives are computed using the Sobel operator. After that, it finds eigenvectors and eigenvalues of \f$M\f$ and stores them in the destination image as \f$(\lambda_1, \lambda_2, x_1, y_1, x_2, y_2)\f$ where

  • \f$\lambda_1, \lambda_2\f$ are the non-sorted eigenvalues of \f$M\f$
  • \f$x_1, y_1\f$ are the eigenvectors corresponding to \f$\lambda_1\f$
  • \f$x_2, y_2\f$ are the eigenvectors corresponding to \f$\lambda_2\f$

The output of the function can be used for robust edge or corner detection.

@sa cornerMinEigenVal, cornerHarris, preCornerDetect

Python prototype (for reference only):

cornerEigenValsAndVecs(src, blockSize, ksize[, dst[, borderType]]) -> dst
Link to this function

cornerEigenValsAndVecs(src, blockSize, ksize, opts)

View Source
@spec cornerEigenValsAndVecs(
  Evision.Mat.maybe_mat_in(),
  integer(),
  integer(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Calculates eigenvalues and eigenvectors of image blocks for corner detection.

Positional Arguments
  • src: Evision.Mat.t().

    Input single-channel 8-bit or floating-point image.

  • blockSize: int.

    Neighborhood size (see details below).

  • ksize: int.

    Aperture parameter for the Sobel operator.

Keyword Arguments
  • borderType: int.

    Pixel extrapolation method. See #BorderTypes. #BORDER_WRAP is not supported.

Return
  • dst: Evision.Mat.t().

    Image to store the results. It has the same size as src and the type CV_32FC(6) .

For every pixel \f$p\f$ , the function cornerEigenValsAndVecs considers a blockSize \f$\times\f$ blockSize neighborhood \f$S(p)\f$ . It calculates the covariation matrix of derivatives over the neighborhood as: \f[M = \begin{bmatrix} \sum _{S(p)}(dI/dx)^2 & \sum _{S(p)}dI/dx dI/dy \\ \sum _{S(p)}dI/dx dI/dy & \sum _{S(p)}(dI/dy)^2 \end{bmatrix}\f] where the derivatives are computed using the Sobel operator. After that, it finds eigenvectors and eigenvalues of \f$M\f$ and stores them in the destination image as \f$(\lambda_1, \lambda_2, x_1, y_1, x_2, y_2)\f$ where

  • \f$\lambda_1, \lambda_2\f$ are the non-sorted eigenvalues of \f$M\f$
  • \f$x_1, y_1\f$ are the eigenvectors corresponding to \f$\lambda_1\f$
  • \f$x_2, y_2\f$ are the eigenvectors corresponding to \f$\lambda_2\f$

The output of the function can be used for robust edge or corner detection.

@sa cornerMinEigenVal, cornerHarris, preCornerDetect

Python prototype (for reference only):

cornerEigenValsAndVecs(src, blockSize, ksize[, dst[, borderType]]) -> dst
Link to this function

cornerHarris(src, blockSize, ksize, k)

View Source
@spec cornerHarris(Evision.Mat.maybe_mat_in(), integer(), integer(), number()) ::
  Evision.Mat.t() | {:error, String.t()}

Harris corner detector.

Positional Arguments
  • src: Evision.Mat.t().

    Input single-channel 8-bit or floating-point image.

  • blockSize: int.

    Neighborhood size (see the details on #cornerEigenValsAndVecs ).

  • ksize: int.

    Aperture parameter for the Sobel operator.

  • k: double.

    Harris detector free parameter. See the formula above.

Keyword Arguments
  • borderType: int.

    Pixel extrapolation method. See #BorderTypes. #BORDER_WRAP is not supported.

Return
  • dst: Evision.Mat.t().

    Image to store the Harris detector responses. It has the type CV_32FC1 and the same size as src .

The function runs the Harris corner detector on the image. Similarly to cornerMinEigenVal and cornerEigenValsAndVecs , for each pixel \f$(x, y)\f$ it calculates a \f$2\times2\f$ gradient covariance matrix \f$M^{(x,y)}\f$ over a \f$\texttt{blockSize} \times \texttt{blockSize}\f$ neighborhood. Then, it computes the following characteristic: \f[\texttt{dst} (x,y) = \mathrm{det} M^{(x,y)} - k \cdot \left ( \mathrm{tr} M^{(x,y)} \right )^2\f] Corners in the image can be found as the local maxima of this response map.

Python prototype (for reference only):

cornerHarris(src, blockSize, ksize, k[, dst[, borderType]]) -> dst
Link to this function

cornerHarris(src, blockSize, ksize, k, opts)

View Source
@spec cornerHarris(
  Evision.Mat.maybe_mat_in(),
  integer(),
  integer(),
  number(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Harris corner detector.

Positional Arguments
  • src: Evision.Mat.t().

    Input single-channel 8-bit or floating-point image.

  • blockSize: int.

    Neighborhood size (see the details on #cornerEigenValsAndVecs ).

  • ksize: int.

    Aperture parameter for the Sobel operator.

  • k: double.

    Harris detector free parameter. See the formula above.

Keyword Arguments
  • borderType: int.

    Pixel extrapolation method. See #BorderTypes. #BORDER_WRAP is not supported.

Return
  • dst: Evision.Mat.t().

    Image to store the Harris detector responses. It has the type CV_32FC1 and the same size as src .

The function runs the Harris corner detector on the image. Similarly to cornerMinEigenVal and cornerEigenValsAndVecs , for each pixel \f$(x, y)\f$ it calculates a \f$2\times2\f$ gradient covariance matrix \f$M^{(x,y)}\f$ over a \f$\texttt{blockSize} \times \texttt{blockSize}\f$ neighborhood. Then, it computes the following characteristic: \f[\texttt{dst} (x,y) = \mathrm{det} M^{(x,y)} - k \cdot \left ( \mathrm{tr} M^{(x,y)} \right )^2\f] Corners in the image can be found as the local maxima of this response map.

Python prototype (for reference only):

cornerHarris(src, blockSize, ksize, k[, dst[, borderType]]) -> dst
Link to this function

cornerMinEigenVal(src, blockSize)

View Source
@spec cornerMinEigenVal(Evision.Mat.maybe_mat_in(), integer()) ::
  Evision.Mat.t() | {:error, String.t()}

Calculates the minimal eigenvalue of gradient matrices for corner detection.

Positional Arguments
  • src: Evision.Mat.t().

    Input single-channel 8-bit or floating-point image.

  • blockSize: int.

    Neighborhood size (see the details on #cornerEigenValsAndVecs ).

Keyword Arguments
  • ksize: int.

    Aperture parameter for the Sobel operator.

  • borderType: int.

    Pixel extrapolation method. See #BorderTypes. #BORDER_WRAP is not supported.

Return
  • dst: Evision.Mat.t().

    Image to store the minimal eigenvalues. It has the type CV_32FC1 and the same size as src .

The function is similar to cornerEigenValsAndVecs but it calculates and stores only the minimal eigenvalue of the covariance matrix of derivatives, that is, \f$\min(\lambda_1, \lambda_2)\f$ in terms of the formulae in the cornerEigenValsAndVecs description.

Python prototype (for reference only):

cornerMinEigenVal(src, blockSize[, dst[, ksize[, borderType]]]) -> dst
Link to this function

cornerMinEigenVal(src, blockSize, opts)

View Source
@spec cornerMinEigenVal(
  Evision.Mat.maybe_mat_in(),
  integer(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Calculates the minimal eigenvalue of gradient matrices for corner detection.

Positional Arguments
  • src: Evision.Mat.t().

    Input single-channel 8-bit or floating-point image.

  • blockSize: int.

    Neighborhood size (see the details on #cornerEigenValsAndVecs ).

Keyword Arguments
  • ksize: int.

    Aperture parameter for the Sobel operator.

  • borderType: int.

    Pixel extrapolation method. See #BorderTypes. #BORDER_WRAP is not supported.

Return
  • dst: Evision.Mat.t().

    Image to store the minimal eigenvalues. It has the type CV_32FC1 and the same size as src .

The function is similar to cornerEigenValsAndVecs but it calculates and stores only the minimal eigenvalue of the covariance matrix of derivatives, that is, \f$\min(\lambda_1, \lambda_2)\f$ in terms of the formulae in the cornerEigenValsAndVecs description.

Python prototype (for reference only):

cornerMinEigenVal(src, blockSize[, dst[, ksize[, borderType]]]) -> dst
Link to this function

cornerSubPix(image, corners, winSize, zeroZone, criteria)

View Source
@spec cornerSubPix(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  {number(), number()},
  {integer(), integer(), number()}
) :: Evision.Mat.t() | {:error, String.t()}

Refines the corner locations.

Positional Arguments
  • image: Evision.Mat.t().

    Input single-channel, 8-bit or float image.

  • winSize: Size.

    Half of the side length of the search window. For example, if winSize=Size(5,5) , then a \f$(52+1) \times (52+1) = 11 \times 11\f$ search window is used.

  • zeroZone: Size.

    Half of the size of the dead region in the middle of the search zone over which the summation in the formula below is not done. It is used sometimes to avoid possible singularities of the autocorrelation matrix. The value of (-1,-1) indicates that there is no such a size.

  • criteria: TermCriteria.

    Criteria for termination of the iterative process of corner refinement. That is, the process of corner position refinement stops either after criteria.maxCount iterations or when the corner position moves by less than criteria.epsilon on some iteration.

Return
  • corners: Evision.Mat.t().

    Initial coordinates of the input corners and refined coordinates provided for output.

The function iterates to find the sub-pixel accurate location of corners or radial saddle points as described in @cite forstner1987fast, and as shown on the figure below. image Sub-pixel accurate corner locator is based on the observation that every vector from the center \f$q\f$ to a point \f$p\f$ located within a neighborhood of \f$q\f$ is orthogonal to the image gradient at \f$p\f$ subject to image and measurement noise. Consider the expression: \f[\epsilon _i = {DI_{p_i}}^T \cdot (q - p_i)\f] where \f${DI_{p_i}}\f$ is an image gradient at one of the points \f$p_i\f$ in a neighborhood of \f$q\f$ . The value of \f$q\f$ is to be found so that \f$\epsilon_i\f$ is minimized. A system of equations may be set up with \f$\epsilon_i\f$ set to zero: \f[\sum _i(DI_{p_i} \cdot {DI_{p_i}}^T) \cdot q - \sum _i(DI_{p_i} \cdot {DI_{p_i}}^T \cdot p_i)\f] where the gradients are summed within a neighborhood ("search window") of \f$q\f$ . Calling the first gradient term \f$G\f$ and the second gradient term \f$b\f$ gives: \f[q = G^{-1} \cdot b\f] The algorithm sets the center of the neighborhood window at this new center \f$q\f$ and then iterates until the center stays within a set threshold.

Python prototype (for reference only):

cornerSubPix(image, corners, winSize, zeroZone, criteria) -> corners
Link to this function

correctMatches(f, points1, points2)

View Source

Refines coordinates of corresponding points.

Positional Arguments
  • f: Evision.Mat.t().

    3x3 fundamental matrix.

  • points1: Evision.Mat.t().

    1xN array containing the first set of points.

  • points2: Evision.Mat.t().

    1xN array containing the second set of points.

Return
  • newPoints1: Evision.Mat.t().

    The optimized points1.

  • newPoints2: Evision.Mat.t().

    The optimized points2.

The function implements the Optimal Triangulation Method (see Multiple View Geometry @cite HartleyZ00 for details). For each given point correspondence points1[i] \<-> points2[i], and a fundamental matrix F, it computes the corrected correspondences newPoints1[i] \<-> newPoints2[i] that minimize the geometric error \f$d(points1[i], newPoints1[i])^2 + d(points2[i],newPoints2[i])^2\f$ (where \f$d(a,b)\f$ is the geometric distance between points \f$a\f$ and \f$b\f$ ) subject to the epipolar constraint \f$newPoints2^T \cdot F \cdot newPoints1 = 0\f$ .

Python prototype (for reference only):

correctMatches(F, points1, points2[, newPoints1[, newPoints2]]) -> newPoints1, newPoints2
Link to this function

correctMatches(f, points1, points2, opts)

View Source
@spec correctMatches(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Refines coordinates of corresponding points.

Positional Arguments
  • f: Evision.Mat.t().

    3x3 fundamental matrix.

  • points1: Evision.Mat.t().

    1xN array containing the first set of points.

  • points2: Evision.Mat.t().

    1xN array containing the second set of points.

Return
  • newPoints1: Evision.Mat.t().

    The optimized points1.

  • newPoints2: Evision.Mat.t().

    The optimized points2.

The function implements the Optimal Triangulation Method (see Multiple View Geometry @cite HartleyZ00 for details). For each given point correspondence points1[i] \<-> points2[i], and a fundamental matrix F, it computes the corrected correspondences newPoints1[i] \<-> newPoints2[i] that minimize the geometric error \f$d(points1[i], newPoints1[i])^2 + d(points2[i],newPoints2[i])^2\f$ (where \f$d(a,b)\f$ is the geometric distance between points \f$a\f$ and \f$b\f$ ) subject to the epipolar constraint \f$newPoints2^T \cdot F \cdot newPoints1 = 0\f$ .

Python prototype (for reference only):

correctMatches(F, points1, points2[, newPoints1[, newPoints2]]) -> newPoints1, newPoints2
@spec countNonZero(Evision.Mat.maybe_mat_in()) :: integer() | {:error, String.t()}

Counts non-zero array elements.

Positional Arguments
  • src: Evision.Mat.t().

    single-channel array.

Return
  • retval: int

The function returns the number of non-zero elements in src : \f[\sum _{I: \; \texttt{src} (I) \ne0 } 1\f] @sa mean, meanStdDev, norm, minMaxLoc, calcCovarMatrix

Python prototype (for reference only):

countNonZero(src) -> retval
Link to this function

createAffineTransformer(fullAffine)

View Source
@spec createAffineTransformer(boolean()) ::
  Evision.AffineTransformer.t() | {:error, String.t()}

createAffineTransformer

Positional Arguments
  • fullAffine: bool
Return
  • retval: Evision.AffineTransformer.t()

Complete constructor

Python prototype (for reference only):

createAffineTransformer(fullAffine) -> retval
@spec createAlignMTB() :: Evision.AlignMTB.t() | {:error, String.t()}

Creates AlignMTB object

Keyword Arguments
  • max_bits: int.

    logarithm to the base 2 of maximal shift in each dimension. Values of 5 and 6 are usually good enough (31 and 63 pixels shift respectively).

  • exclude_range: int.

    range for exclusion bitmap that is constructed to suppress noise around the median value.

  • cut: bool.

    if true cuts images, otherwise fills the new regions with zeros.

Return
  • retval: Evision.AlignMTB.t()

Python prototype (for reference only):

createAlignMTB([, max_bits[, exclude_range[, cut]]]) -> retval
@spec createAlignMTB([{atom(), term()}, ...] | nil) ::
  Evision.AlignMTB.t() | {:error, String.t()}

Creates AlignMTB object

Keyword Arguments
  • max_bits: int.

    logarithm to the base 2 of maximal shift in each dimension. Values of 5 and 6 are usually good enough (31 and 63 pixels shift respectively).

  • exclude_range: int.

    range for exclusion bitmap that is constructed to suppress noise around the median value.

  • cut: bool.

    if true cuts images, otherwise fills the new regions with zeros.

Return
  • retval: Evision.AlignMTB.t()

Python prototype (for reference only):

createAlignMTB([, max_bits[, exclude_range[, cut]]]) -> retval
Link to this function

createBackgroundSubtractorKNN()

View Source
@spec createBackgroundSubtractorKNN() ::
  Evision.BackgroundSubtractorKNN.t() | {:error, String.t()}

Creates KNN Background Subtractor

Keyword Arguments
  • history: int.

    Length of the history.

  • dist2Threshold: double.

    Threshold on the squared distance between the pixel and the sample to decide whether a pixel is close to that sample. This parameter does not affect the background update.

  • detectShadows: bool.

    If true, the algorithm will detect shadows and mark them. It decreases the speed a bit, so if you do not need this feature, set the parameter to false.

Return
  • retval: Evision.BackgroundSubtractorKNN.t()

Python prototype (for reference only):

createBackgroundSubtractorKNN([, history[, dist2Threshold[, detectShadows]]]) -> retval
Link to this function

createBackgroundSubtractorKNN(opts)

View Source
@spec createBackgroundSubtractorKNN([{atom(), term()}, ...] | nil) ::
  Evision.BackgroundSubtractorKNN.t() | {:error, String.t()}

Creates KNN Background Subtractor

Keyword Arguments
  • history: int.

    Length of the history.

  • dist2Threshold: double.

    Threshold on the squared distance between the pixel and the sample to decide whether a pixel is close to that sample. This parameter does not affect the background update.

  • detectShadows: bool.

    If true, the algorithm will detect shadows and mark them. It decreases the speed a bit, so if you do not need this feature, set the parameter to false.

Return
  • retval: Evision.BackgroundSubtractorKNN.t()

Python prototype (for reference only):

createBackgroundSubtractorKNN([, history[, dist2Threshold[, detectShadows]]]) -> retval
Link to this function

createBackgroundSubtractorMOG2()

View Source
@spec createBackgroundSubtractorMOG2() ::
  Evision.BackgroundSubtractorMOG2.t() | {:error, String.t()}

Creates MOG2 Background Subtractor

Keyword Arguments
  • history: int.

    Length of the history.

  • varThreshold: double.

    Threshold on the squared Mahalanobis distance between the pixel and the model to decide whether a pixel is well described by the background model. This parameter does not affect the background update.

  • detectShadows: bool.

    If true, the algorithm will detect shadows and mark them. It decreases the speed a bit, so if you do not need this feature, set the parameter to false.

Return
  • retval: Evision.BackgroundSubtractorMOG2.t()

Python prototype (for reference only):

createBackgroundSubtractorMOG2([, history[, varThreshold[, detectShadows]]]) -> retval
Link to this function

createBackgroundSubtractorMOG2(opts)

View Source
@spec createBackgroundSubtractorMOG2([{atom(), term()}, ...] | nil) ::
  Evision.BackgroundSubtractorMOG2.t() | {:error, String.t()}

Creates MOG2 Background Subtractor

Keyword Arguments
  • history: int.

    Length of the history.

  • varThreshold: double.

    Threshold on the squared Mahalanobis distance between the pixel and the model to decide whether a pixel is well described by the background model. This parameter does not affect the background update.

  • detectShadows: bool.

    If true, the algorithm will detect shadows and mark them. It decreases the speed a bit, so if you do not need this feature, set the parameter to false.

Return
  • retval: Evision.BackgroundSubtractorMOG2.t()

Python prototype (for reference only):

createBackgroundSubtractorMOG2([, history[, varThreshold[, detectShadows]]]) -> retval
Link to this function

createCalibrateDebevec()

View Source
@spec createCalibrateDebevec() :: Evision.CalibrateDebevec.t() | {:error, String.t()}

Creates CalibrateDebevec object

Keyword Arguments
  • samples: int.

    number of pixel locations to use

  • lambda: float.

    smoothness term weight. Greater values produce smoother results, but can alter the response.

  • random: bool.

    if true sample pixel locations are chosen at random, otherwise they form a rectangular grid.

Return
  • retval: Evision.CalibrateDebevec.t()

Python prototype (for reference only):

createCalibrateDebevec([, samples[, lambda[, random]]]) -> retval
Link to this function

createCalibrateDebevec(opts)

View Source
@spec createCalibrateDebevec([{atom(), term()}, ...] | nil) ::
  Evision.CalibrateDebevec.t() | {:error, String.t()}

Creates CalibrateDebevec object

Keyword Arguments
  • samples: int.

    number of pixel locations to use

  • lambda: float.

    smoothness term weight. Greater values produce smoother results, but can alter the response.

  • random: bool.

    if true sample pixel locations are chosen at random, otherwise they form a rectangular grid.

Return
  • retval: Evision.CalibrateDebevec.t()

Python prototype (for reference only):

createCalibrateDebevec([, samples[, lambda[, random]]]) -> retval
Link to this function

createCalibrateRobertson()

View Source
@spec createCalibrateRobertson() ::
  Evision.CalibrateRobertson.t() | {:error, String.t()}

Creates CalibrateRobertson object

Keyword Arguments
  • max_iter: int.

    maximal number of Gauss-Seidel solver iterations.

  • threshold: float.

    target difference between results of two successive steps of the minimization.

Return
  • retval: Evision.CalibrateRobertson.t()

Python prototype (for reference only):

createCalibrateRobertson([, max_iter[, threshold]]) -> retval
Link to this function

createCalibrateRobertson(opts)

View Source
@spec createCalibrateRobertson([{atom(), term()}, ...] | nil) ::
  Evision.CalibrateRobertson.t() | {:error, String.t()}

Creates CalibrateRobertson object

Keyword Arguments
  • max_iter: int.

    maximal number of Gauss-Seidel solver iterations.

  • threshold: float.

    target difference between results of two successive steps of the minimization.

Return
  • retval: Evision.CalibrateRobertson.t()

Python prototype (for reference only):

createCalibrateRobertson([, max_iter[, threshold]]) -> retval
Link to this function

createChiHistogramCostExtractor()

View Source
@spec createChiHistogramCostExtractor() ::
  Evision.HistogramCostExtractor.t() | {:error, String.t()}

createChiHistogramCostExtractor

Keyword Arguments
  • nDummies: int.
  • defaultCost: float.
Return
  • retval: Evision.HistogramCostExtractor.t()

Python prototype (for reference only):

createChiHistogramCostExtractor([, nDummies[, defaultCost]]) -> retval
Link to this function

createChiHistogramCostExtractor(opts)

View Source
@spec createChiHistogramCostExtractor([{atom(), term()}, ...] | nil) ::
  Evision.HistogramCostExtractor.t() | {:error, String.t()}

createChiHistogramCostExtractor

Keyword Arguments
  • nDummies: int.
  • defaultCost: float.
Return
  • retval: Evision.HistogramCostExtractor.t()

Python prototype (for reference only):

createChiHistogramCostExtractor([, nDummies[, defaultCost]]) -> retval
@spec createCLAHE() :: Evision.CLAHE.t() | {:error, String.t()}

Creates a smart pointer to a cv::CLAHE class and initializes it.

Keyword Arguments
  • clipLimit: double.

    Threshold for contrast limiting.

  • tileGridSize: Size.

    Size of grid for histogram equalization. Input image will be divided into equally sized rectangular tiles. tileGridSize defines the number of tiles in row and column.

Return
  • retval: Evision.CLAHE.t()

Python prototype (for reference only):

createCLAHE([, clipLimit[, tileGridSize]]) -> retval
@spec createCLAHE([{atom(), term()}, ...] | nil) ::
  Evision.CLAHE.t() | {:error, String.t()}

Creates a smart pointer to a cv::CLAHE class and initializes it.

Keyword Arguments
  • clipLimit: double.

    Threshold for contrast limiting.

  • tileGridSize: Size.

    Size of grid for histogram equalization. Input image will be divided into equally sized rectangular tiles. tileGridSize defines the number of tiles in row and column.

Return
  • retval: Evision.CLAHE.t()

Python prototype (for reference only):

createCLAHE([, clipLimit[, tileGridSize]]) -> retval
Link to this function

createEMDHistogramCostExtractor()

View Source
@spec createEMDHistogramCostExtractor() ::
  Evision.HistogramCostExtractor.t() | {:error, String.t()}

createEMDHistogramCostExtractor

Keyword Arguments
  • flag: int.
  • nDummies: int.
  • defaultCost: float.
Return
  • retval: Evision.HistogramCostExtractor.t()

Python prototype (for reference only):

createEMDHistogramCostExtractor([, flag[, nDummies[, defaultCost]]]) -> retval
Link to this function

createEMDHistogramCostExtractor(opts)

View Source
@spec createEMDHistogramCostExtractor([{atom(), term()}, ...] | nil) ::
  Evision.HistogramCostExtractor.t() | {:error, String.t()}

createEMDHistogramCostExtractor

Keyword Arguments
  • flag: int.
  • nDummies: int.
  • defaultCost: float.
Return
  • retval: Evision.HistogramCostExtractor.t()

Python prototype (for reference only):

createEMDHistogramCostExtractor([, flag[, nDummies[, defaultCost]]]) -> retval
Link to this function

createEMDL1HistogramCostExtractor()

View Source
@spec createEMDL1HistogramCostExtractor() ::
  Evision.HistogramCostExtractor.t() | {:error, String.t()}

createEMDL1HistogramCostExtractor

Keyword Arguments
  • nDummies: int.
  • defaultCost: float.
Return
  • retval: Evision.HistogramCostExtractor.t()

Python prototype (for reference only):

createEMDL1HistogramCostExtractor([, nDummies[, defaultCost]]) -> retval
Link to this function

createEMDL1HistogramCostExtractor(opts)

View Source
@spec createEMDL1HistogramCostExtractor([{atom(), term()}, ...] | nil) ::
  Evision.HistogramCostExtractor.t() | {:error, String.t()}

createEMDL1HistogramCostExtractor

Keyword Arguments
  • nDummies: int.
  • defaultCost: float.
Return
  • retval: Evision.HistogramCostExtractor.t()

Python prototype (for reference only):

createEMDL1HistogramCostExtractor([, nDummies[, defaultCost]]) -> retval
Link to this function

createGeneralizedHoughBallard()

View Source
@spec createGeneralizedHoughBallard() ::
  Evision.GeneralizedHoughBallard.t() | {:error, String.t()}

Creates a smart pointer to a cv::GeneralizedHoughBallard class and initializes it.

Return
  • retval: Evision.GeneralizedHoughBallard.t()

Python prototype (for reference only):

createGeneralizedHoughBallard() -> retval
Link to this function

createGeneralizedHoughGuil()

View Source
@spec createGeneralizedHoughGuil() ::
  Evision.GeneralizedHoughGuil.t() | {:error, String.t()}

Creates a smart pointer to a cv::GeneralizedHoughGuil class and initializes it.

Return
  • retval: Evision.GeneralizedHoughGuil.t()

Python prototype (for reference only):

createGeneralizedHoughGuil() -> retval
Link to this function

createHanningWindow(winSize, type)

View Source
@spec createHanningWindow(
  {number(), number()},
  integer()
) :: Evision.Mat.t() | {:error, String.t()}

This function computes a Hanning window coefficients in two dimensions.

Positional Arguments
  • winSize: Size.

    The window size specifications (both width and height must be > 1)

  • type: int.

    Created array type

Return
  • dst: Evision.Mat.t().

    Destination array to place Hann coefficients in

See (http://en.wikipedia.org/wiki/Hann_function) and (http://en.wikipedia.org/wiki/Window_function) for more information. An example is shown below:

// create hanning window of size 100x100 and type CV_32F
Mat hann;
createHanningWindow(hann, Size(100, 100), CV_32F);

Python prototype (for reference only):

createHanningWindow(winSize, type[, dst]) -> dst
Link to this function

createHanningWindow(winSize, type, opts)

View Source
@spec createHanningWindow(
  {number(), number()},
  integer(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

This function computes a Hanning window coefficients in two dimensions.

Positional Arguments
  • winSize: Size.

    The window size specifications (both width and height must be > 1)

  • type: int.

    Created array type

Return
  • dst: Evision.Mat.t().

    Destination array to place Hann coefficients in

See (http://en.wikipedia.org/wiki/Hann_function) and (http://en.wikipedia.org/wiki/Window_function) for more information. An example is shown below:

// create hanning window of size 100x100 and type CV_32F
Mat hann;
createHanningWindow(hann, Size(100, 100), CV_32F);

Python prototype (for reference only):

createHanningWindow(winSize, type[, dst]) -> dst
Link to this function

createHausdorffDistanceExtractor()

View Source
@spec createHausdorffDistanceExtractor() ::
  Evision.HausdorffDistanceExtractor.t() | {:error, String.t()}

createHausdorffDistanceExtractor

Keyword Arguments
  • distanceFlag: int.
  • rankProp: float.
Return
  • retval: Evision.HausdorffDistanceExtractor.t()

Python prototype (for reference only):

createHausdorffDistanceExtractor([, distanceFlag[, rankProp]]) -> retval
Link to this function

createHausdorffDistanceExtractor(opts)

View Source
@spec createHausdorffDistanceExtractor([{atom(), term()}, ...] | nil) ::
  Evision.HausdorffDistanceExtractor.t() | {:error, String.t()}

createHausdorffDistanceExtractor

Keyword Arguments
  • distanceFlag: int.
  • rankProp: float.
Return
  • retval: Evision.HausdorffDistanceExtractor.t()

Python prototype (for reference only):

createHausdorffDistanceExtractor([, distanceFlag[, rankProp]]) -> retval
Link to this function

createLineSegmentDetector()

View Source
@spec createLineSegmentDetector() ::
  Evision.LineSegmentDetector.t() | {:error, String.t()}

Creates a smart pointer to a LineSegmentDetector object and initializes it.

Keyword Arguments
  • refine: int.

    The way found lines will be refined, see #LineSegmentDetectorModes

  • scale: double.

    The scale of the image that will be used to find the lines. Range (0..1].

  • sigma_scale: double.

    Sigma for Gaussian filter. It is computed as sigma = sigma_scale/scale.

  • quant: double.

    Bound to the quantization error on the gradient norm.

  • ang_th: double.

    Gradient angle tolerance in degrees.

  • log_eps: double.

    Detection threshold: -log10(NFA) > log_eps. Used only when advance refinement is chosen.

  • density_th: double.

    Minimal density of aligned region points in the enclosing rectangle.

  • n_bins: int.

    Number of bins in pseudo-ordering of gradient modulus.

Return
  • retval: Evision.LineSegmentDetector.t()

The LineSegmentDetector algorithm is defined using the standard values. Only advanced users may want to edit those, as to tailor it for their own application.

Python prototype (for reference only):

createLineSegmentDetector([, refine[, scale[, sigma_scale[, quant[, ang_th[, log_eps[, density_th[, n_bins]]]]]]]]) -> retval
Link to this function

createLineSegmentDetector(opts)

View Source
@spec createLineSegmentDetector([{atom(), term()}, ...] | nil) ::
  Evision.LineSegmentDetector.t() | {:error, String.t()}

Creates a smart pointer to a LineSegmentDetector object and initializes it.

Keyword Arguments
  • refine: int.

    The way found lines will be refined, see #LineSegmentDetectorModes

  • scale: double.

    The scale of the image that will be used to find the lines. Range (0..1].

  • sigma_scale: double.

    Sigma for Gaussian filter. It is computed as sigma = sigma_scale/scale.

  • quant: double.

    Bound to the quantization error on the gradient norm.

  • ang_th: double.

    Gradient angle tolerance in degrees.

  • log_eps: double.

    Detection threshold: -log10(NFA) > log_eps. Used only when advance refinement is chosen.

  • density_th: double.

    Minimal density of aligned region points in the enclosing rectangle.

  • n_bins: int.

    Number of bins in pseudo-ordering of gradient modulus.

Return
  • retval: Evision.LineSegmentDetector.t()

The LineSegmentDetector algorithm is defined using the standard values. Only advanced users may want to edit those, as to tailor it for their own application.

Python prototype (for reference only):

createLineSegmentDetector([, refine[, scale[, sigma_scale[, quant[, ang_th[, log_eps[, density_th[, n_bins]]]]]]]]) -> retval
@spec createMergeDebevec() :: Evision.MergeDebevec.t() | {:error, String.t()}

Creates MergeDebevec object

Return
  • retval: Evision.MergeDebevec.t()

Python prototype (for reference only):

createMergeDebevec() -> retval
@spec createMergeMertens() :: Evision.MergeMertens.t() | {:error, String.t()}

Creates MergeMertens object

Keyword Arguments
  • contrast_weight: float.

    contrast measure weight. See MergeMertens.

  • saturation_weight: float.

    saturation measure weight

  • exposure_weight: float.

    well-exposedness measure weight

Return
  • retval: Evision.MergeMertens.t()

Python prototype (for reference only):

createMergeMertens([, contrast_weight[, saturation_weight[, exposure_weight]]]) -> retval
Link to this function

createMergeMertens(opts)

View Source
@spec createMergeMertens([{atom(), term()}, ...] | nil) ::
  Evision.MergeMertens.t() | {:error, String.t()}

Creates MergeMertens object

Keyword Arguments
  • contrast_weight: float.

    contrast measure weight. See MergeMertens.

  • saturation_weight: float.

    saturation measure weight

  • exposure_weight: float.

    well-exposedness measure weight

Return
  • retval: Evision.MergeMertens.t()

Python prototype (for reference only):

createMergeMertens([, contrast_weight[, saturation_weight[, exposure_weight]]]) -> retval
@spec createMergeRobertson() :: Evision.MergeRobertson.t() | {:error, String.t()}

Creates MergeRobertson object

Return
  • retval: Evision.MergeRobertson.t()

Python prototype (for reference only):

createMergeRobertson() -> retval
Link to this function

createNormHistogramCostExtractor()

View Source
@spec createNormHistogramCostExtractor() ::
  Evision.HistogramCostExtractor.t() | {:error, String.t()}

createNormHistogramCostExtractor

Keyword Arguments
  • flag: int.
  • nDummies: int.
  • defaultCost: float.
Return
  • retval: Evision.HistogramCostExtractor.t()

Python prototype (for reference only):

createNormHistogramCostExtractor([, flag[, nDummies[, defaultCost]]]) -> retval
Link to this function

createNormHistogramCostExtractor(opts)

View Source
@spec createNormHistogramCostExtractor([{atom(), term()}, ...] | nil) ::
  Evision.HistogramCostExtractor.t() | {:error, String.t()}

createNormHistogramCostExtractor

Keyword Arguments
  • flag: int.
  • nDummies: int.
  • defaultCost: float.
Return
  • retval: Evision.HistogramCostExtractor.t()

Python prototype (for reference only):

createNormHistogramCostExtractor([, flag[, nDummies[, defaultCost]]]) -> retval
Link to this function

createShapeContextDistanceExtractor()

View Source
@spec createShapeContextDistanceExtractor() ::
  Evision.ShapeContextDistanceExtractor.t() | {:error, String.t()}

createShapeContextDistanceExtractor

Keyword Arguments
  • nAngularBins: int.
  • nRadialBins: int.
  • innerRadius: float.
  • outerRadius: float.
  • iterations: int.
  • comparer: Evision.HistogramCostExtractor.t().
  • transformer: Evision.ShapeTransformer.t().
Return
  • retval: Evision.ShapeContextDistanceExtractor.t()

Python prototype (for reference only):

createShapeContextDistanceExtractor([, nAngularBins[, nRadialBins[, innerRadius[, outerRadius[, iterations[, comparer[, transformer]]]]]]]) -> retval
Link to this function

createShapeContextDistanceExtractor(opts)

View Source
@spec createShapeContextDistanceExtractor([{atom(), term()}, ...] | nil) ::
  Evision.ShapeContextDistanceExtractor.t() | {:error, String.t()}

createShapeContextDistanceExtractor

Keyword Arguments
  • nAngularBins: int.
  • nRadialBins: int.
  • innerRadius: float.
  • outerRadius: float.
  • iterations: int.
  • comparer: Evision.HistogramCostExtractor.t().
  • transformer: Evision.ShapeTransformer.t().
Return
  • retval: Evision.ShapeContextDistanceExtractor.t()

Python prototype (for reference only):

createShapeContextDistanceExtractor([, nAngularBins[, nRadialBins[, innerRadius[, outerRadius[, iterations[, comparer[, transformer]]]]]]]) -> retval
Link to this function

createThinPlateSplineShapeTransformer()

View Source
@spec createThinPlateSplineShapeTransformer() ::
  Evision.ThinPlateSplineShapeTransformer.t() | {:error, String.t()}

createThinPlateSplineShapeTransformer

Keyword Arguments
  • regularizationParameter: double.
Return
  • retval: Evision.ThinPlateSplineShapeTransformer.t()

Complete constructor

Python prototype (for reference only):

createThinPlateSplineShapeTransformer([, regularizationParameter]) -> retval
Link to this function

createThinPlateSplineShapeTransformer(opts)

View Source
@spec createThinPlateSplineShapeTransformer([{atom(), term()}, ...] | nil) ::
  Evision.ThinPlateSplineShapeTransformer.t() | {:error, String.t()}

createThinPlateSplineShapeTransformer

Keyword Arguments
  • regularizationParameter: double.
Return
  • retval: Evision.ThinPlateSplineShapeTransformer.t()

Complete constructor

Python prototype (for reference only):

createThinPlateSplineShapeTransformer([, regularizationParameter]) -> retval
@spec createTonemap() :: Evision.Tonemap.t() | {:error, String.t()}

Creates simple linear mapper with gamma correction

Keyword Arguments
  • gamma: float.

    positive value for gamma correction. Gamma value of 1.0 implies no correction, gamma equal to 2.2f is suitable for most displays. Generally gamma > 1 brightens the image and gamma \< 1 darkens it.

Return
  • retval: Evision.Tonemap.t()

Python prototype (for reference only):

createTonemap([, gamma]) -> retval
@spec createTonemap([{atom(), term()}, ...] | nil) ::
  Evision.Tonemap.t() | {:error, String.t()}

Creates simple linear mapper with gamma correction

Keyword Arguments
  • gamma: float.

    positive value for gamma correction. Gamma value of 1.0 implies no correction, gamma equal to 2.2f is suitable for most displays. Generally gamma > 1 brightens the image and gamma \< 1 darkens it.

Return
  • retval: Evision.Tonemap.t()

Python prototype (for reference only):

createTonemap([, gamma]) -> retval
@spec createTonemapDrago() :: Evision.TonemapDrago.t() | {:error, String.t()}

Creates TonemapDrago object

Keyword Arguments
  • gamma: float.

    gamma value for gamma correction. See createTonemap

  • saturation: float.

    positive saturation enhancement value. 1.0 preserves saturation, values greater than 1 increase saturation and values less than 1 decrease it.

  • bias: float.

    value for bias function in [0, 1] range. Values from 0.7 to 0.9 usually give best results, default value is 0.85.

Return
  • retval: Evision.TonemapDrago.t()

Python prototype (for reference only):

createTonemapDrago([, gamma[, saturation[, bias]]]) -> retval
Link to this function

createTonemapDrago(opts)

View Source
@spec createTonemapDrago([{atom(), term()}, ...] | nil) ::
  Evision.TonemapDrago.t() | {:error, String.t()}

Creates TonemapDrago object

Keyword Arguments
  • gamma: float.

    gamma value for gamma correction. See createTonemap

  • saturation: float.

    positive saturation enhancement value. 1.0 preserves saturation, values greater than 1 increase saturation and values less than 1 decrease it.

  • bias: float.

    value for bias function in [0, 1] range. Values from 0.7 to 0.9 usually give best results, default value is 0.85.

Return
  • retval: Evision.TonemapDrago.t()

Python prototype (for reference only):

createTonemapDrago([, gamma[, saturation[, bias]]]) -> retval
@spec createTonemapMantiuk() :: Evision.TonemapMantiuk.t() | {:error, String.t()}

Creates TonemapMantiuk object

Keyword Arguments
  • gamma: float.

    gamma value for gamma correction. See createTonemap

  • scale: float.

    contrast scale factor. HVS response is multiplied by this parameter, thus compressing dynamic range. Values from 0.6 to 0.9 produce best results.

  • saturation: float.

    saturation enhancement value. See createTonemapDrago

Return
  • retval: Evision.TonemapMantiuk.t()

Python prototype (for reference only):

createTonemapMantiuk([, gamma[, scale[, saturation]]]) -> retval
Link to this function

createTonemapMantiuk(opts)

View Source
@spec createTonemapMantiuk([{atom(), term()}, ...] | nil) ::
  Evision.TonemapMantiuk.t() | {:error, String.t()}

Creates TonemapMantiuk object

Keyword Arguments
  • gamma: float.

    gamma value for gamma correction. See createTonemap

  • scale: float.

    contrast scale factor. HVS response is multiplied by this parameter, thus compressing dynamic range. Values from 0.6 to 0.9 produce best results.

  • saturation: float.

    saturation enhancement value. See createTonemapDrago

Return
  • retval: Evision.TonemapMantiuk.t()

Python prototype (for reference only):

createTonemapMantiuk([, gamma[, scale[, saturation]]]) -> retval
@spec createTonemapReinhard() :: Evision.TonemapReinhard.t() | {:error, String.t()}

Creates TonemapReinhard object

Keyword Arguments
  • gamma: float.

    gamma value for gamma correction. See createTonemap

  • intensity: float.

    result intensity in [-8, 8] range. Greater intensity produces brighter results.

  • light_adapt: float.

    light adaptation in [0, 1] range. If 1 adaptation is based only on pixel value, if 0 it's global, otherwise it's a weighted mean of this two cases.

  • color_adapt: float.

    chromatic adaptation in [0, 1] range. If 1 channels are treated independently, if 0 adaptation level is the same for each channel.

Return
  • retval: Evision.TonemapReinhard.t()

Python prototype (for reference only):

createTonemapReinhard([, gamma[, intensity[, light_adapt[, color_adapt]]]]) -> retval
Link to this function

createTonemapReinhard(opts)

View Source
@spec createTonemapReinhard([{atom(), term()}, ...] | nil) ::
  Evision.TonemapReinhard.t() | {:error, String.t()}

Creates TonemapReinhard object

Keyword Arguments
  • gamma: float.

    gamma value for gamma correction. See createTonemap

  • intensity: float.

    result intensity in [-8, 8] range. Greater intensity produces brighter results.

  • light_adapt: float.

    light adaptation in [0, 1] range. If 1 adaptation is based only on pixel value, if 0 it's global, otherwise it's a weighted mean of this two cases.

  • color_adapt: float.

    chromatic adaptation in [0, 1] range. If 1 channels are treated independently, if 0 adaptation level is the same for each channel.

Return
  • retval: Evision.TonemapReinhard.t()

Python prototype (for reference only):

createTonemapReinhard([, gamma[, intensity[, light_adapt[, color_adapt]]]]) -> retval
@spec cubeRoot(number()) :: number() | {:error, String.t()}

Computes the cube root of an argument.

Positional Arguments
  • val: float.

    A function argument.

Return
  • retval: float

The function cubeRoot computes \f$\sqrt[3]{\texttt{val}}\f$. Negative arguments are handled correctly. NaN and Inf are not handled. The accuracy approaches the maximum possible accuracy for single-precision data.

Python prototype (for reference only):

cubeRoot(val) -> retval
@spec cvtColor(Evision.Mat.maybe_mat_in(), integer()) ::
  Evision.Mat.t() | {:error, String.t()}

Converts an image from one color space to another.

Positional Arguments
  • src: Evision.Mat.t().

    input image: 8-bit unsigned, 16-bit unsigned ( CV_16UC... ), or single-precision floating-point.

  • code: int.

    color space conversion code (see #ColorConversionCodes).

Keyword Arguments
  • dstCn: int.

    number of channels in the destination image; if the parameter is 0, the number of the channels is derived automatically from src and code.

Return
  • dst: Evision.Mat.t().

    output image of the same size and depth as src.

The function converts an input image from one color space to another. In case of a transformation to-from RGB color space, the order of the channels should be specified explicitly (RGB or BGR). Note that the default color format in OpenCV is often referred to as RGB but it is actually BGR (the bytes are reversed). So the first byte in a standard (24-bit) color image will be an 8-bit Blue component, the second byte will be Green, and the third byte will be Red. The fourth, fifth, and sixth bytes would then be the second pixel (Blue, then Green, then Red), and so on. The conventional ranges for R, G, and B channel values are:

  • 0 to 255 for CV_8U images
  • 0 to 65535 for CV_16U images
  • 0 to 1 for CV_32F images

In case of linear transformations, the range does not matter. But in case of a non-linear transformation, an input RGB image should be normalized to the proper value range to get the correct results, for example, for RGB \f$\rightarrow\f$ L*u*v* transformation. For example, if you have a 32-bit floating-point image directly converted from an 8-bit image without any scaling, then it will have the 0..255 value range instead of 0..1 assumed by the function. So, before calling #cvtColor , you need first to scale the image down:

img *= 1./255;
cvtColor(img, img, COLOR_BGR2Luv);

If you use #cvtColor with 8-bit images, the conversion will have some information lost. For many applications, this will not be noticeable but it is recommended to use 32-bit images in applications that need the full range of colors or that convert an image before an operation and then convert back. If conversion adds the alpha channel, its value will set to the maximum of corresponding channel range: 255 for CV_8U, 65535 for CV_16U, 1 for CV_32F.

@see @ref imgproc_color_conversions

Python prototype (for reference only):

cvtColor(src, code[, dst[, dstCn]]) -> dst
Link to this function

cvtColor(src, code, opts)

View Source
@spec cvtColor(Evision.Mat.maybe_mat_in(), integer(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Converts an image from one color space to another.

Positional Arguments
  • src: Evision.Mat.t().

    input image: 8-bit unsigned, 16-bit unsigned ( CV_16UC... ), or single-precision floating-point.

  • code: int.

    color space conversion code (see #ColorConversionCodes).

Keyword Arguments
  • dstCn: int.

    number of channels in the destination image; if the parameter is 0, the number of the channels is derived automatically from src and code.

Return
  • dst: Evision.Mat.t().

    output image of the same size and depth as src.

The function converts an input image from one color space to another. In case of a transformation to-from RGB color space, the order of the channels should be specified explicitly (RGB or BGR). Note that the default color format in OpenCV is often referred to as RGB but it is actually BGR (the bytes are reversed). So the first byte in a standard (24-bit) color image will be an 8-bit Blue component, the second byte will be Green, and the third byte will be Red. The fourth, fifth, and sixth bytes would then be the second pixel (Blue, then Green, then Red), and so on. The conventional ranges for R, G, and B channel values are:

  • 0 to 255 for CV_8U images
  • 0 to 65535 for CV_16U images
  • 0 to 1 for CV_32F images

In case of linear transformations, the range does not matter. But in case of a non-linear transformation, an input RGB image should be normalized to the proper value range to get the correct results, for example, for RGB \f$\rightarrow\f$ L*u*v* transformation. For example, if you have a 32-bit floating-point image directly converted from an 8-bit image without any scaling, then it will have the 0..255 value range instead of 0..1 assumed by the function. So, before calling #cvtColor , you need first to scale the image down:

img *= 1./255;
cvtColor(img, img, COLOR_BGR2Luv);

If you use #cvtColor with 8-bit images, the conversion will have some information lost. For many applications, this will not be noticeable but it is recommended to use 32-bit images in applications that need the full range of colors or that convert an image before an operation and then convert back. If conversion adds the alpha channel, its value will set to the maximum of corresponding channel range: 255 for CV_8U, 65535 for CV_16U, 1 for CV_32F.

@see @ref imgproc_color_conversions

Python prototype (for reference only):

cvtColor(src, code[, dst[, dstCn]]) -> dst
Link to this function

cvtColorTwoPlane(src1, src2, code)

View Source
@spec cvtColorTwoPlane(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  integer()
) ::
  Evision.Mat.t() | {:error, String.t()}

Converts an image from one color space to another where the source image is stored in two planes.

Positional Arguments
  • src1: Evision.Mat.t().

    8-bit image (#CV_8U) of the Y plane.

  • src2: Evision.Mat.t().

    image containing interleaved U/V plane.

  • code: int.

    Specifies the type of conversion. It can take any of the following values:

    • #COLOR_YUV2BGR_NV12
    • #COLOR_YUV2RGB_NV12
    • #COLOR_YUV2BGRA_NV12
    • #COLOR_YUV2RGBA_NV12
    • #COLOR_YUV2BGR_NV21
    • #COLOR_YUV2RGB_NV21
    • #COLOR_YUV2BGRA_NV21
    • #COLOR_YUV2RGBA_NV21
Return
  • dst: Evision.Mat.t().

    output image.

This function only supports YUV420 to RGB conversion as of now.

Python prototype (for reference only):

cvtColorTwoPlane(src1, src2, code[, dst]) -> dst
Link to this function

cvtColorTwoPlane(src1, src2, code, opts)

View Source
@spec cvtColorTwoPlane(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  integer(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Converts an image from one color space to another where the source image is stored in two planes.

Positional Arguments
  • src1: Evision.Mat.t().

    8-bit image (#CV_8U) of the Y plane.

  • src2: Evision.Mat.t().

    image containing interleaved U/V plane.

  • code: int.

    Specifies the type of conversion. It can take any of the following values:

    • #COLOR_YUV2BGR_NV12
    • #COLOR_YUV2RGB_NV12
    • #COLOR_YUV2BGRA_NV12
    • #COLOR_YUV2RGBA_NV12
    • #COLOR_YUV2BGR_NV21
    • #COLOR_YUV2RGB_NV21
    • #COLOR_YUV2BGRA_NV21
    • #COLOR_YUV2RGBA_NV21
Return
  • dst: Evision.Mat.t().

    output image.

This function only supports YUV420 to RGB conversion as of now.

Python prototype (for reference only):

cvtColorTwoPlane(src1, src2, code[, dst]) -> dst
@spec dct(Evision.Mat.maybe_mat_in()) :: Evision.Mat.t() | {:error, String.t()}

Performs a forward or inverse discrete Cosine transform of 1D or 2D array.

Positional Arguments
  • src: Evision.Mat.t().

    input floating-point array.

Keyword Arguments
  • flags: int.

    transformation flags as a combination of cv::DftFlags (DCT_*)

Return
  • dst: Evision.Mat.t().

    output array of the same size and type as src .

The function cv::dct performs a forward or inverse discrete Cosine transform (DCT) of a 1D or 2D floating-point array:

  • Forward Cosine transform of a 1D vector of N elements: \f[Y = C^{(N)} \cdot X\f] where \f[C^{(N)}_{jk}= \sqrt{\alpha_j/N} \cos \left ( \frac{\pi(2k+1)j}{2N} \right )\f] and \f$\alpha_0=1\f$, \f$\alpha_j=2\f$ for j > 0.

  • Inverse Cosine transform of a 1D vector of N elements: \f[X = \left (C^{(N)} \right )^{-1} \cdot Y = \left (C^{(N)} \right )^T \cdot Y\f] (since \f$C^{(N)}\f$ is an orthogonal matrix, \f$C^{(N)} \cdot \left(C^{(N)}\right)^T = I\f$ )

  • Forward 2D Cosine transform of M x N matrix: \f[Y = C^{(N)} \cdot X \cdot \left (C^{(N)} \right )^T\f]

  • Inverse 2D Cosine transform of M x N matrix: \f[X = \left (C^{(N)} \right )^T \cdot X \cdot C^{(N)}\f]

The function chooses the mode of operation by looking at the flags and size of the input array:

  • If (flags & #DCT_INVERSE) == 0 , the function does a forward 1D or 2D transform. Otherwise, it is an inverse 1D or 2D transform.

  • If (flags & #DCT_ROWS) != 0 , the function performs a 1D transform of each row.

  • If the array is a single column or a single row, the function performs a 1D transform.

  • If none of the above is true, the function performs a 2D transform.

Note: Currently dct supports even-size arrays (2, 4, 6 ...). For data analysis and approximation, you can pad the array when necessary. Also, the function performance depends very much, and not monotonically, on the array size (see getOptimalDFTSize ). In the current implementation DCT of a vector of size N is calculated via DFT of a vector of size N/2 . Thus, the optimal DCT size N1 >= N can be calculated as:

size_t getOptimalDCTSize(size_t N) { return 2*getOptimalDFTSize((N+1)/2); }
N1 = getOptimalDCTSize(N);

@sa dft , getOptimalDFTSize , idct

Python prototype (for reference only):

dct(src[, dst[, flags]]) -> dst
@spec dct(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Performs a forward or inverse discrete Cosine transform of 1D or 2D array.

Positional Arguments
  • src: Evision.Mat.t().

    input floating-point array.

Keyword Arguments
  • flags: int.

    transformation flags as a combination of cv::DftFlags (DCT_*)

Return
  • dst: Evision.Mat.t().

    output array of the same size and type as src .

The function cv::dct performs a forward or inverse discrete Cosine transform (DCT) of a 1D or 2D floating-point array:

  • Forward Cosine transform of a 1D vector of N elements: \f[Y = C^{(N)} \cdot X\f] where \f[C^{(N)}_{jk}= \sqrt{\alpha_j/N} \cos \left ( \frac{\pi(2k+1)j}{2N} \right )\f] and \f$\alpha_0=1\f$, \f$\alpha_j=2\f$ for j > 0.

  • Inverse Cosine transform of a 1D vector of N elements: \f[X = \left (C^{(N)} \right )^{-1} \cdot Y = \left (C^{(N)} \right )^T \cdot Y\f] (since \f$C^{(N)}\f$ is an orthogonal matrix, \f$C^{(N)} \cdot \left(C^{(N)}\right)^T = I\f$ )

  • Forward 2D Cosine transform of M x N matrix: \f[Y = C^{(N)} \cdot X \cdot \left (C^{(N)} \right )^T\f]

  • Inverse 2D Cosine transform of M x N matrix: \f[X = \left (C^{(N)} \right )^T \cdot X \cdot C^{(N)}\f]

The function chooses the mode of operation by looking at the flags and size of the input array:

  • If (flags & #DCT_INVERSE) == 0 , the function does a forward 1D or 2D transform. Otherwise, it is an inverse 1D or 2D transform.

  • If (flags & #DCT_ROWS) != 0 , the function performs a 1D transform of each row.

  • If the array is a single column or a single row, the function performs a 1D transform.

  • If none of the above is true, the function performs a 2D transform.

Note: Currently dct supports even-size arrays (2, 4, 6 ...). For data analysis and approximation, you can pad the array when necessary. Also, the function performance depends very much, and not monotonically, on the array size (see getOptimalDFTSize ). In the current implementation DCT of a vector of size N is calculated via DFT of a vector of size N/2 . Thus, the optimal DCT size N1 >= N can be calculated as:

size_t getOptimalDCTSize(size_t N) { return 2*getOptimalDFTSize((N+1)/2); }
N1 = getOptimalDCTSize(N);

@sa dft , getOptimalDFTSize , idct

Python prototype (for reference only):

dct(src[, dst[, flags]]) -> dst
@spec decolor(Evision.Mat.maybe_mat_in()) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Transforms a color image to a grayscale image. It is a basic tool in digital printing, stylized black-and-white photograph rendering, and in many single channel image processing applications

Positional Arguments
  • src: Evision.Mat.t().

    Input 8-bit 3-channel image.

Return
  • grayscale: Evision.Mat.t().

    Output 8-bit 1-channel image.

  • color_boost: Evision.Mat.t().

    Output 8-bit 3-channel image.

@cite CL12 .

This function is to be applied on color images.

Python prototype (for reference only):

decolor(src[, grayscale[, color_boost]]) -> grayscale, color_boost
@spec decolor(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Transforms a color image to a grayscale image. It is a basic tool in digital printing, stylized black-and-white photograph rendering, and in many single channel image processing applications

Positional Arguments
  • src: Evision.Mat.t().

    Input 8-bit 3-channel image.

Return
  • grayscale: Evision.Mat.t().

    Output 8-bit 1-channel image.

  • color_boost: Evision.Mat.t().

    Output 8-bit 3-channel image.

@cite CL12 .

This function is to be applied on color images.

Python prototype (for reference only):

decolor(src[, grayscale[, color_boost]]) -> grayscale, color_boost
Link to this function

decomposeEssentialMat(e)

View Source
@spec decomposeEssentialMat(Evision.Mat.maybe_mat_in()) ::
  {Evision.Mat.t(), Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Decompose an essential matrix to possible rotations and translation.

Positional Arguments
  • e: Evision.Mat.t().

    The input essential matrix.

Return
  • r1: Evision.Mat.t().

    One possible rotation matrix.

  • r2: Evision.Mat.t().

    Another possible rotation matrix.

  • t: Evision.Mat.t().

    One possible translation.

This function decomposes the essential matrix E using svd decomposition @cite HartleyZ00. In general, four possible poses exist for the decomposition of E. They are \f$[R_1, t]\f$, \f$[R_1, -t]\f$, \f$[R_2, t]\f$, \f$[R_2, -t]\f$. If E gives the epipolar constraint \f$[p_2; 1]^T A^{-T} E A^{-1} [p_1; 1] = 0\f$ between the image points \f$p_1\f$ in the first image and \f$p_2\f$ in second image, then any of the tuples \f$[R_1, t]\f$, \f$[R_1, -t]\f$, \f$[R_2, t]\f$, \f$[R_2, -t]\f$ is a change of basis from the first camera's coordinate system to the second camera's coordinate system. However, by decomposing E, one can only get the direction of the translation. For this reason, the translation t is returned with unit length.

Python prototype (for reference only):

decomposeEssentialMat(E[, R1[, R2[, t]]]) -> R1, R2, t
Link to this function

decomposeEssentialMat(e, opts)

View Source
@spec decomposeEssentialMat(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  {Evision.Mat.t(), Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Decompose an essential matrix to possible rotations and translation.

Positional Arguments
  • e: Evision.Mat.t().

    The input essential matrix.

Return
  • r1: Evision.Mat.t().

    One possible rotation matrix.

  • r2: Evision.Mat.t().

    Another possible rotation matrix.

  • t: Evision.Mat.t().

    One possible translation.

This function decomposes the essential matrix E using svd decomposition @cite HartleyZ00. In general, four possible poses exist for the decomposition of E. They are \f$[R_1, t]\f$, \f$[R_1, -t]\f$, \f$[R_2, t]\f$, \f$[R_2, -t]\f$. If E gives the epipolar constraint \f$[p_2; 1]^T A^{-T} E A^{-1} [p_1; 1] = 0\f$ between the image points \f$p_1\f$ in the first image and \f$p_2\f$ in second image, then any of the tuples \f$[R_1, t]\f$, \f$[R_1, -t]\f$, \f$[R_2, t]\f$, \f$[R_2, -t]\f$ is a change of basis from the first camera's coordinate system to the second camera's coordinate system. However, by decomposing E, one can only get the direction of the translation. For this reason, the translation t is returned with unit length.

Python prototype (for reference only):

decomposeEssentialMat(E[, R1[, R2[, t]]]) -> R1, R2, t
Link to this function

decomposeHomographyMat(h, k)

View Source
@spec decomposeHomographyMat(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in()) ::
  {integer(), [Evision.Mat.t()], [Evision.Mat.t()], [Evision.Mat.t()]}
  | {:error, String.t()}

Decompose a homography matrix to rotation(s), translation(s) and plane normal(s).

Positional Arguments
  • h: Evision.Mat.t().

    The input homography matrix between two images.

  • k: Evision.Mat.t().

    The input camera intrinsic matrix.

Return
  • retval: int

  • rotations: [Evision.Mat].

    Array of rotation matrices.

  • translations: [Evision.Mat].

    Array of translation matrices.

  • normals: [Evision.Mat].

    Array of plane normal matrices.

This function extracts relative camera motion between two views of a planar object and returns up to four mathematical solution tuples of rotation, translation, and plane normal. The decomposition of the homography matrix H is described in detail in @cite Malis2007. If the homography H, induced by the plane, gives the constraint \f[s_i \vecthree{x'_i}{y'_i}{1} \sim H \vecthree{x_i}{y_i}{1}\f] on the source image points \f$p_i\f$ and the destination image points \f$p'_i\f$, then the tuple of rotations[k] and translations[k] is a change of basis from the source camera's coordinate system to the destination camera's coordinate system. However, by decomposing H, one can only get the translation normalized by the (typically unknown) depth of the scene, i.e. its direction but with normalized length. If point correspondences are available, at least two solutions may further be invalidated, by applying positive depth constraint, i.e. all points must be in front of the camera.

Python prototype (for reference only):

decomposeHomographyMat(H, K[, rotations[, translations[, normals]]]) -> retval, rotations, translations, normals
Link to this function

decomposeHomographyMat(h, k, opts)

View Source
@spec decomposeHomographyMat(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  {integer(), [Evision.Mat.t()], [Evision.Mat.t()], [Evision.Mat.t()]}
  | {:error, String.t()}

Decompose a homography matrix to rotation(s), translation(s) and plane normal(s).

Positional Arguments
  • h: Evision.Mat.t().

    The input homography matrix between two images.

  • k: Evision.Mat.t().

    The input camera intrinsic matrix.

Return
  • retval: int

  • rotations: [Evision.Mat].

    Array of rotation matrices.

  • translations: [Evision.Mat].

    Array of translation matrices.

  • normals: [Evision.Mat].

    Array of plane normal matrices.

This function extracts relative camera motion between two views of a planar object and returns up to four mathematical solution tuples of rotation, translation, and plane normal. The decomposition of the homography matrix H is described in detail in @cite Malis2007. If the homography H, induced by the plane, gives the constraint \f[s_i \vecthree{x'_i}{y'_i}{1} \sim H \vecthree{x_i}{y_i}{1}\f] on the source image points \f$p_i\f$ and the destination image points \f$p'_i\f$, then the tuple of rotations[k] and translations[k] is a change of basis from the source camera's coordinate system to the destination camera's coordinate system. However, by decomposing H, one can only get the translation normalized by the (typically unknown) depth of the scene, i.e. its direction but with normalized length. If point correspondences are available, at least two solutions may further be invalidated, by applying positive depth constraint, i.e. all points must be in front of the camera.

Python prototype (for reference only):

decomposeHomographyMat(H, K[, rotations[, translations[, normals]]]) -> retval, rotations, translations, normals
Link to this function

decomposeProjectionMatrix(projMatrix)

View Source
@spec decomposeProjectionMatrix(Evision.Mat.maybe_mat_in()) ::
  {Evision.Mat.t(), Evision.Mat.t(), Evision.Mat.t(), Evision.Mat.t(),
   Evision.Mat.t(), Evision.Mat.t(), Evision.Mat.t()}
  | {:error, String.t()}

Decomposes a projection matrix into a rotation matrix and a camera intrinsic matrix.

Positional Arguments
  • projMatrix: Evision.Mat.t().

    3x4 input projection matrix P.

Return
  • cameraMatrix: Evision.Mat.t().

    Output 3x3 camera intrinsic matrix \f$\cameramatrix{A}\f$.

  • rotMatrix: Evision.Mat.t().

    Output 3x3 external rotation matrix R.

  • transVect: Evision.Mat.t().

    Output 4x1 translation vector T.

  • rotMatrixX: Evision.Mat.t().

    Optional 3x3 rotation matrix around x-axis.

  • rotMatrixY: Evision.Mat.t().

    Optional 3x3 rotation matrix around y-axis.

  • rotMatrixZ: Evision.Mat.t().

    Optional 3x3 rotation matrix around z-axis.

  • eulerAngles: Evision.Mat.t().

    Optional three-element vector containing three Euler angles of rotation in degrees.

The function computes a decomposition of a projection matrix into a calibration and a rotation matrix and the position of a camera. It optionally returns three rotation matrices, one for each axis, and three Euler angles that could be used in OpenGL. Note, there is always more than one sequence of rotations about the three principal axes that results in the same orientation of an object, e.g. see @cite Slabaugh . Returned three rotation matrices and corresponding three Euler angles are only one of the possible solutions. The function is based on #RQDecomp3x3 .

Python prototype (for reference only):

decomposeProjectionMatrix(projMatrix[, cameraMatrix[, rotMatrix[, transVect[, rotMatrixX[, rotMatrixY[, rotMatrixZ[, eulerAngles]]]]]]]) -> cameraMatrix, rotMatrix, transVect, rotMatrixX, rotMatrixY, rotMatrixZ, eulerAngles
Link to this function

decomposeProjectionMatrix(projMatrix, opts)

View Source
@spec decomposeProjectionMatrix(
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  {Evision.Mat.t(), Evision.Mat.t(), Evision.Mat.t(), Evision.Mat.t(),
   Evision.Mat.t(), Evision.Mat.t(), Evision.Mat.t()}
  | {:error, String.t()}

Decomposes a projection matrix into a rotation matrix and a camera intrinsic matrix.

Positional Arguments
  • projMatrix: Evision.Mat.t().

    3x4 input projection matrix P.

Return
  • cameraMatrix: Evision.Mat.t().

    Output 3x3 camera intrinsic matrix \f$\cameramatrix{A}\f$.

  • rotMatrix: Evision.Mat.t().

    Output 3x3 external rotation matrix R.

  • transVect: Evision.Mat.t().

    Output 4x1 translation vector T.

  • rotMatrixX: Evision.Mat.t().

    Optional 3x3 rotation matrix around x-axis.

  • rotMatrixY: Evision.Mat.t().

    Optional 3x3 rotation matrix around y-axis.

  • rotMatrixZ: Evision.Mat.t().

    Optional 3x3 rotation matrix around z-axis.

  • eulerAngles: Evision.Mat.t().

    Optional three-element vector containing three Euler angles of rotation in degrees.

The function computes a decomposition of a projection matrix into a calibration and a rotation matrix and the position of a camera. It optionally returns three rotation matrices, one for each axis, and three Euler angles that could be used in OpenGL. Note, there is always more than one sequence of rotations about the three principal axes that results in the same orientation of an object, e.g. see @cite Slabaugh . Returned three rotation matrices and corresponding three Euler angles are only one of the possible solutions. The function is based on #RQDecomp3x3 .

Python prototype (for reference only):

decomposeProjectionMatrix(projMatrix[, cameraMatrix[, rotMatrix[, transVect[, rotMatrixX[, rotMatrixY[, rotMatrixZ[, eulerAngles]]]]]]]) -> cameraMatrix, rotMatrix, transVect, rotMatrixX, rotMatrixY, rotMatrixZ, eulerAngles
@spec demosaicing(Evision.Mat.maybe_mat_in(), integer()) ::
  Evision.Mat.t() | {:error, String.t()}

main function for all demosaicing processes

Positional Arguments
  • src: Evision.Mat.t().

    input image: 8-bit unsigned or 16-bit unsigned.

  • code: int.

    Color space conversion code (see the description below).

Keyword Arguments
  • dstCn: int.

    number of channels in the destination image; if the parameter is 0, the number of the channels is derived automatically from src and code.

Return
  • dst: Evision.Mat.t().

    output image of the same size and depth as src.

The function can do the following transformations:

  • Demosaicing using bilinear interpolation

#COLOR_BayerBG2BGR , #COLOR_BayerGB2BGR , #COLOR_BayerRG2BGR , #COLOR_BayerGR2BGR #COLOR_BayerBG2GRAY , #COLOR_BayerGB2GRAY , #COLOR_BayerRG2GRAY , #COLOR_BayerGR2GRAY

  • Demosaicing using Variable Number of Gradients.

#COLOR_BayerBG2BGR_VNG , #COLOR_BayerGB2BGR_VNG , #COLOR_BayerRG2BGR_VNG , #COLOR_BayerGR2BGR_VNG

  • Edge-Aware Demosaicing.

#COLOR_BayerBG2BGR_EA , #COLOR_BayerGB2BGR_EA , #COLOR_BayerRG2BGR_EA , #COLOR_BayerGR2BGR_EA

  • Demosaicing with alpha channel

#COLOR_BayerBG2BGRA , #COLOR_BayerGB2BGRA , #COLOR_BayerRG2BGRA , #COLOR_BayerGR2BGRA @sa cvtColor

Python prototype (for reference only):

demosaicing(src, code[, dst[, dstCn]]) -> dst
Link to this function

demosaicing(src, code, opts)

View Source
@spec demosaicing(
  Evision.Mat.maybe_mat_in(),
  integer(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

main function for all demosaicing processes

Positional Arguments
  • src: Evision.Mat.t().

    input image: 8-bit unsigned or 16-bit unsigned.

  • code: int.

    Color space conversion code (see the description below).

Keyword Arguments
  • dstCn: int.

    number of channels in the destination image; if the parameter is 0, the number of the channels is derived automatically from src and code.

Return
  • dst: Evision.Mat.t().

    output image of the same size and depth as src.

The function can do the following transformations:

  • Demosaicing using bilinear interpolation

#COLOR_BayerBG2BGR , #COLOR_BayerGB2BGR , #COLOR_BayerRG2BGR , #COLOR_BayerGR2BGR #COLOR_BayerBG2GRAY , #COLOR_BayerGB2GRAY , #COLOR_BayerRG2GRAY , #COLOR_BayerGR2GRAY

  • Demosaicing using Variable Number of Gradients.

#COLOR_BayerBG2BGR_VNG , #COLOR_BayerGB2BGR_VNG , #COLOR_BayerRG2BGR_VNG , #COLOR_BayerGR2BGR_VNG

  • Edge-Aware Demosaicing.

#COLOR_BayerBG2BGR_EA , #COLOR_BayerGB2BGR_EA , #COLOR_BayerRG2BGR_EA , #COLOR_BayerGR2BGR_EA

  • Demosaicing with alpha channel

#COLOR_BayerBG2BGRA , #COLOR_BayerGB2BGRA , #COLOR_BayerRG2BGRA , #COLOR_BayerGR2BGRA @sa cvtColor

Python prototype (for reference only):

demosaicing(src, code[, dst[, dstCn]]) -> dst
@spec detailEnhance(Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

This filter enhances the details of a particular image.

Positional Arguments
  • src: Evision.Mat.t().

    Input 8-bit 3-channel image.

Keyword Arguments
  • sigma_s: float.

    %Range between 0 to 200.

  • sigma_r: float.

    %Range between 0 to 1.

Return
  • dst: Evision.Mat.t().

    Output image with the same size and type as src.

Python prototype (for reference only):

detailEnhance(src[, dst[, sigma_s[, sigma_r]]]) -> dst
Link to this function

detailEnhance(src, opts)

View Source
@spec detailEnhance(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

This filter enhances the details of a particular image.

Positional Arguments
  • src: Evision.Mat.t().

    Input 8-bit 3-channel image.

Keyword Arguments
  • sigma_s: float.

    %Range between 0 to 200.

  • sigma_r: float.

    %Range between 0 to 1.

Return
  • dst: Evision.Mat.t().

    Output image with the same size and type as src.

Python prototype (for reference only):

detailEnhance(src[, dst[, sigma_s[, sigma_r]]]) -> dst
@spec determinant(Evision.Mat.maybe_mat_in()) :: number() | {:error, String.t()}

Returns the determinant of a square floating-point matrix.

Positional Arguments
  • mtx: Evision.Mat.t().

    input matrix that must have CV_32FC1 or CV_64FC1 type and square size.

Return
  • retval: double

The function cv::determinant calculates and returns the determinant of the specified matrix. For small matrices ( mtx.cols=mtx.rows\<=3 ), the direct method is used. For larger matrices, the function uses LU factorization with partial pivoting. For symmetric positively-determined matrices, it is also possible to use eigen decomposition to calculate the determinant. @sa trace, invert, solve, eigen, @ref MatrixExpressions

Python prototype (for reference only):

determinant(mtx) -> retval
@spec dft(Evision.Mat.maybe_mat_in()) :: Evision.Mat.t() | {:error, String.t()}

Performs a forward or inverse Discrete Fourier transform of a 1D or 2D floating-point array.

Positional Arguments
  • src: Evision.Mat.t().

    input array that could be real or complex.

Keyword Arguments
  • flags: int.

    transformation flags, representing a combination of the #DftFlags

  • nonzeroRows: int.

    when the parameter is not zero, the function assumes that only the first nonzeroRows rows of the input array (#DFT_INVERSE is not set) or only the first nonzeroRows of the output array (#DFT_INVERSE is set) contain non-zeros, thus, the function can handle the rest of the rows more efficiently and save some time; this technique is very useful for calculating array cross-correlation or convolution using DFT.

Return
  • dst: Evision.Mat.t().

    output array whose size and type depends on the flags .

The function cv::dft performs one of the following:

  • Forward the Fourier transform of a 1D vector of N elements: \f[Y = F^{(N)} \cdot X,\f] where \f$F^{(N)}_{jk}=\exp(-2\pi i j k/N)\f$ and \f$i=\sqrt{-1}\f$

  • Inverse the Fourier transform of a 1D vector of N elements: \f[\begin{array}{l} X'= \left (F^{(N)} \right )^{-1} \cdot Y = \left (F^{(N)} \right )^* \cdot y \\ X = (1/N) \cdot X, \end{array}\f] where \f$F^*=\left(\textrm{Re}(F^{(N)})-\textrm{Im}(F^{(N)})\right)^T\f$

  • Forward the 2D Fourier transform of a M x N matrix: \f[Y = F^{(M)} \cdot X \cdot F^{(N)}\f]

  • Inverse the 2D Fourier transform of a M x N matrix: \f[\begin{array}{l} X'= \left (F^{(M)} \right )^* \cdot Y \cdot \left (F^{(N)} \right )^* \\ X = \frac{1}{M \cdot N} \cdot X' \end{array}\f]

In case of real (single-channel) data, the output spectrum of the forward Fourier transform or input spectrum of the inverse Fourier transform can be represented in a packed format called CCS (complex-conjugate-symmetrical). It was borrowed from IPL (Intel* Image Processing Library). Here is how 2D CCS spectrum looks: \f[\begin{bmatrix} Re Y_{0,0} & Re Y_{0,1} & Im Y_{0,1} & Re Y_{0,2} & Im Y_{0,2} & \cdots & Re Y_{0,N/2-1} & Im Y_{0,N/2-1} & Re Y_{0,N/2} \\ Re Y_{1,0} & Re Y_{1,1} & Im Y_{1,1} & Re Y_{1,2} & Im Y_{1,2} & \cdots & Re Y_{1,N/2-1} & Im Y_{1,N/2-1} & Re Y_{1,N/2} \\ Im Y_{1,0} & Re Y_{2,1} & Im Y_{2,1} & Re Y_{2,2} & Im Y_{2,2} & \cdots & Re Y_{2,N/2-1} & Im Y_{2,N/2-1} & Im Y_{1,N/2} \\ \hdotsfor{9} \\ Re Y_{M/2-1,0} & Re Y_{M-3,1} & Im Y_{M-3,1} & \hdotsfor{3} & Re Y_{M-3,N/2-1} & Im Y_{M-3,N/2-1}& Re Y_{M/2-1,N/2} \\ Im Y_{M/2-1,0} & Re Y_{M-2,1} & Im Y_{M-2,1} & \hdotsfor{3} & Re Y_{M-2,N/2-1} & Im Y_{M-2,N/2-1}& Im Y_{M/2-1,N/2} \\ Re Y_{M/2,0} & Re Y_{M-1,1} & Im Y_{M-1,1} & \hdotsfor{3} & Re Y_{M-1,N/2-1} & Im Y_{M-1,N/2-1}& Re Y_{M/2,N/2} \end{bmatrix}\f] In case of 1D transform of a real vector, the output looks like the first row of the matrix above. So, the function chooses an operation mode depending on the flags and size of the input array:

  • If #DFT_ROWS is set or the input array has a single row or single column, the function performs a 1D forward or inverse transform of each row of a matrix when #DFT_ROWS is set. Otherwise, it performs a 2D transform.

  • If the input array is real and #DFT_INVERSE is not set, the function performs a forward 1D or 2D transform:

  • When #DFT_COMPLEX_OUTPUT is set, the output is a complex matrix of the same size as input.

  • When #DFT_COMPLEX_OUTPUT is not set, the output is a real matrix of the same size as input. In case of 2D transform, it uses the packed format as shown above. In case of a single 1D transform, it looks like the first row of the matrix above. In case of multiple 1D transforms (when using the #DFT_ROWS flag), each row of the output matrix looks like the first row of the matrix above.

  • If the input array is complex and either #DFT_INVERSE or #DFT_REAL_OUTPUT are not set, the output is a complex array of the same size as input. The function performs a forward or inverse 1D or 2D transform of the whole input array or each row of the input array independently, depending on the flags DFT_INVERSE and DFT_ROWS.

  • When #DFT_INVERSE is set and the input array is real, or it is complex but #DFT_REAL_OUTPUT is set, the output is a real array of the same size as input. The function performs a 1D or 2D inverse transformation of the whole input array or each individual row, depending on the flags #DFT_INVERSE and #DFT_ROWS.

If #DFT_SCALE is set, the scaling is done after the transformation. Unlike dct , the function supports arrays of arbitrary size. But only those arrays are processed efficiently, whose sizes can be factorized in a product of small prime numbers (2, 3, and 5 in the current implementation). Such an efficient DFT size can be calculated using the getOptimalDFTSize method. The sample below illustrates how to calculate a DFT-based convolution of two 2D real arrays:

void convolveDFT(InputArray A, InputArray B, OutputArray C)
{
// reallocate the output array if needed
C.create(abs(A.rows - B.rows)+1, abs(A.cols - B.cols)+1, A.type());
Size dftSize;
// calculate the size of DFT transform
dftSize.width = getOptimalDFTSize(A.cols + B.cols - 1);
dftSize.height = getOptimalDFTSize(A.rows + B.rows - 1);
// allocate temporary buffers and initialize them with 0's
Mat tempA(dftSize, A.type(), Scalar::all(0));
Mat tempB(dftSize, B.type(), Scalar::all(0));
// copy A and B to the top-left corners of tempA and tempB, respectively
Mat roiA(tempA, Rect(0,0,A.cols,A.rows));
A.copyTo(roiA);
Mat roiB(tempB, Rect(0,0,B.cols,B.rows));
B.copyTo(roiB);
// now transform the padded A & B in-place;
// use "nonzeroRows" hint for faster processing
dft(tempA, tempA, 0, A.rows);
dft(tempB, tempB, 0, B.rows);
// multiply the spectrums;
// the function handles packed spectrum representations well
mulSpectrums(tempA, tempB, tempA);
// transform the product back from the frequency domain.
// Even though all the result rows will be non-zero,
// you need only the first C.rows of them, and thus you
// pass nonzeroRows == C.rows
dft(tempA, tempA, DFT_INVERSE + DFT_SCALE, C.rows);
// now copy the result back to C.
tempA(Rect(0, 0, C.cols, C.rows)).copyTo(C);
// all the temporary buffers will be deallocated automatically
}

To optimize this sample, consider the following approaches:

  • Since nonzeroRows != 0 is passed to the forward transform calls and since A and B are copied to the top-left corners of tempA and tempB, respectively, it is not necessary to clear the whole tempA and tempB. It is only necessary to clear the tempA.cols - A.cols ( tempB.cols - B.cols) rightmost columns of the matrices.

  • This DFT-based convolution does not have to be applied to the whole big arrays, especially if B is significantly smaller than A or vice versa. Instead, you can calculate convolution by parts. To do this, you need to split the output array C into multiple tiles. For each tile, estimate which parts of A and B are required to calculate convolution in this tile. If the tiles in C are too small, the speed will decrease a lot because of repeated work. In the ultimate case, when each tile in C is a single pixel, the algorithm becomes equivalent to the naive convolution algorithm. If the tiles are too big, the temporary arrays tempA and tempB become too big and there is also a slowdown because of bad cache locality. So, there is an optimal tile size somewhere in the middle.

  • If different tiles in C can be calculated in parallel and, thus, the convolution is done by parts, the loop can be threaded.

All of the above improvements have been implemented in #matchTemplate and #filter2D . Therefore, by using them, you can get the performance even better than with the above theoretically optimal implementation. Though, those two functions actually calculate cross-correlation, not convolution, so you need to "flip" the second convolution operand B vertically and horizontally using flip . Note:

  • An example using the discrete fourier transform can be found at opencv_source_code/samples/cpp/dft.cpp

  • (Python) An example using the dft functionality to perform Wiener deconvolution can be found at opencv_source/samples/python/deconvolution.py

  • (Python) An example rearranging the quadrants of a Fourier image can be found at opencv_source/samples/python/dft.py

@sa dct , getOptimalDFTSize , mulSpectrums, filter2D , matchTemplate , flip , cartToPolar , magnitude , phase

Python prototype (for reference only):

dft(src[, dst[, flags[, nonzeroRows]]]) -> dst
@spec dft(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Performs a forward or inverse Discrete Fourier transform of a 1D or 2D floating-point array.

Positional Arguments
  • src: Evision.Mat.t().

    input array that could be real or complex.

Keyword Arguments
  • flags: int.

    transformation flags, representing a combination of the #DftFlags

  • nonzeroRows: int.

    when the parameter is not zero, the function assumes that only the first nonzeroRows rows of the input array (#DFT_INVERSE is not set) or only the first nonzeroRows of the output array (#DFT_INVERSE is set) contain non-zeros, thus, the function can handle the rest of the rows more efficiently and save some time; this technique is very useful for calculating array cross-correlation or convolution using DFT.

Return
  • dst: Evision.Mat.t().

    output array whose size and type depends on the flags .

The function cv::dft performs one of the following:

  • Forward the Fourier transform of a 1D vector of N elements: \f[Y = F^{(N)} \cdot X,\f] where \f$F^{(N)}_{jk}=\exp(-2\pi i j k/N)\f$ and \f$i=\sqrt{-1}\f$

  • Inverse the Fourier transform of a 1D vector of N elements: \f[\begin{array}{l} X'= \left (F^{(N)} \right )^{-1} \cdot Y = \left (F^{(N)} \right )^* \cdot y \\ X = (1/N) \cdot X, \end{array}\f] where \f$F^*=\left(\textrm{Re}(F^{(N)})-\textrm{Im}(F^{(N)})\right)^T\f$

  • Forward the 2D Fourier transform of a M x N matrix: \f[Y = F^{(M)} \cdot X \cdot F^{(N)}\f]

  • Inverse the 2D Fourier transform of a M x N matrix: \f[\begin{array}{l} X'= \left (F^{(M)} \right )^* \cdot Y \cdot \left (F^{(N)} \right )^* \\ X = \frac{1}{M \cdot N} \cdot X' \end{array}\f]

In case of real (single-channel) data, the output spectrum of the forward Fourier transform or input spectrum of the inverse Fourier transform can be represented in a packed format called CCS (complex-conjugate-symmetrical). It was borrowed from IPL (Intel* Image Processing Library). Here is how 2D CCS spectrum looks: \f[\begin{bmatrix} Re Y_{0,0} & Re Y_{0,1} & Im Y_{0,1} & Re Y_{0,2} & Im Y_{0,2} & \cdots & Re Y_{0,N/2-1} & Im Y_{0,N/2-1} & Re Y_{0,N/2} \\ Re Y_{1,0} & Re Y_{1,1} & Im Y_{1,1} & Re Y_{1,2} & Im Y_{1,2} & \cdots & Re Y_{1,N/2-1} & Im Y_{1,N/2-1} & Re Y_{1,N/2} \\ Im Y_{1,0} & Re Y_{2,1} & Im Y_{2,1} & Re Y_{2,2} & Im Y_{2,2} & \cdots & Re Y_{2,N/2-1} & Im Y_{2,N/2-1} & Im Y_{1,N/2} \\ \hdotsfor{9} \\ Re Y_{M/2-1,0} & Re Y_{M-3,1} & Im Y_{M-3,1} & \hdotsfor{3} & Re Y_{M-3,N/2-1} & Im Y_{M-3,N/2-1}& Re Y_{M/2-1,N/2} \\ Im Y_{M/2-1,0} & Re Y_{M-2,1} & Im Y_{M-2,1} & \hdotsfor{3} & Re Y_{M-2,N/2-1} & Im Y_{M-2,N/2-1}& Im Y_{M/2-1,N/2} \\ Re Y_{M/2,0} & Re Y_{M-1,1} & Im Y_{M-1,1} & \hdotsfor{3} & Re Y_{M-1,N/2-1} & Im Y_{M-1,N/2-1}& Re Y_{M/2,N/2} \end{bmatrix}\f] In case of 1D transform of a real vector, the output looks like the first row of the matrix above. So, the function chooses an operation mode depending on the flags and size of the input array:

  • If #DFT_ROWS is set or the input array has a single row or single column, the function performs a 1D forward or inverse transform of each row of a matrix when #DFT_ROWS is set. Otherwise, it performs a 2D transform.

  • If the input array is real and #DFT_INVERSE is not set, the function performs a forward 1D or 2D transform:

  • When #DFT_COMPLEX_OUTPUT is set, the output is a complex matrix of the same size as input.

  • When #DFT_COMPLEX_OUTPUT is not set, the output is a real matrix of the same size as input. In case of 2D transform, it uses the packed format as shown above. In case of a single 1D transform, it looks like the first row of the matrix above. In case of multiple 1D transforms (when using the #DFT_ROWS flag), each row of the output matrix looks like the first row of the matrix above.

  • If the input array is complex and either #DFT_INVERSE or #DFT_REAL_OUTPUT are not set, the output is a complex array of the same size as input. The function performs a forward or inverse 1D or 2D transform of the whole input array or each row of the input array independently, depending on the flags DFT_INVERSE and DFT_ROWS.

  • When #DFT_INVERSE is set and the input array is real, or it is complex but #DFT_REAL_OUTPUT is set, the output is a real array of the same size as input. The function performs a 1D or 2D inverse transformation of the whole input array or each individual row, depending on the flags #DFT_INVERSE and #DFT_ROWS.

If #DFT_SCALE is set, the scaling is done after the transformation. Unlike dct , the function supports arrays of arbitrary size. But only those arrays are processed efficiently, whose sizes can be factorized in a product of small prime numbers (2, 3, and 5 in the current implementation). Such an efficient DFT size can be calculated using the getOptimalDFTSize method. The sample below illustrates how to calculate a DFT-based convolution of two 2D real arrays:

void convolveDFT(InputArray A, InputArray B, OutputArray C)
{
// reallocate the output array if needed
C.create(abs(A.rows - B.rows)+1, abs(A.cols - B.cols)+1, A.type());
Size dftSize;
// calculate the size of DFT transform
dftSize.width = getOptimalDFTSize(A.cols + B.cols - 1);
dftSize.height = getOptimalDFTSize(A.rows + B.rows - 1);
// allocate temporary buffers and initialize them with 0's
Mat tempA(dftSize, A.type(), Scalar::all(0));
Mat tempB(dftSize, B.type(), Scalar::all(0));
// copy A and B to the top-left corners of tempA and tempB, respectively
Mat roiA(tempA, Rect(0,0,A.cols,A.rows));
A.copyTo(roiA);
Mat roiB(tempB, Rect(0,0,B.cols,B.rows));
B.copyTo(roiB);
// now transform the padded A & B in-place;
// use "nonzeroRows" hint for faster processing
dft(tempA, tempA, 0, A.rows);
dft(tempB, tempB, 0, B.rows);
// multiply the spectrums;
// the function handles packed spectrum representations well
mulSpectrums(tempA, tempB, tempA);
// transform the product back from the frequency domain.
// Even though all the result rows will be non-zero,
// you need only the first C.rows of them, and thus you
// pass nonzeroRows == C.rows
dft(tempA, tempA, DFT_INVERSE + DFT_SCALE, C.rows);
// now copy the result back to C.
tempA(Rect(0, 0, C.cols, C.rows)).copyTo(C);
// all the temporary buffers will be deallocated automatically
}

To optimize this sample, consider the following approaches:

  • Since nonzeroRows != 0 is passed to the forward transform calls and since A and B are copied to the top-left corners of tempA and tempB, respectively, it is not necessary to clear the whole tempA and tempB. It is only necessary to clear the tempA.cols - A.cols ( tempB.cols - B.cols) rightmost columns of the matrices.

  • This DFT-based convolution does not have to be applied to the whole big arrays, especially if B is significantly smaller than A or vice versa. Instead, you can calculate convolution by parts. To do this, you need to split the output array C into multiple tiles. For each tile, estimate which parts of A and B are required to calculate convolution in this tile. If the tiles in C are too small, the speed will decrease a lot because of repeated work. In the ultimate case, when each tile in C is a single pixel, the algorithm becomes equivalent to the naive convolution algorithm. If the tiles are too big, the temporary arrays tempA and tempB become too big and there is also a slowdown because of bad cache locality. So, there is an optimal tile size somewhere in the middle.

  • If different tiles in C can be calculated in parallel and, thus, the convolution is done by parts, the loop can be threaded.

All of the above improvements have been implemented in #matchTemplate and #filter2D . Therefore, by using them, you can get the performance even better than with the above theoretically optimal implementation. Though, those two functions actually calculate cross-correlation, not convolution, so you need to "flip" the second convolution operand B vertically and horizontally using flip . Note:

  • An example using the discrete fourier transform can be found at opencv_source_code/samples/cpp/dft.cpp

  • (Python) An example using the dft functionality to perform Wiener deconvolution can be found at opencv_source/samples/python/deconvolution.py

  • (Python) An example rearranging the quadrants of a Fourier image can be found at opencv_source/samples/python/dft.py

@sa dct , getOptimalDFTSize , mulSpectrums, filter2D , matchTemplate , flip , cartToPolar , magnitude , phase

Python prototype (for reference only):

dft(src[, dst[, flags[, nonzeroRows]]]) -> dst

Dilates an image by using a specific structuring element.

Positional Arguments
  • src: Evision.Mat.t().

    input image; the number of channels can be arbitrary, but the depth should be one of CV_8U, CV_16U, CV_16S, CV_32F or CV_64F.

  • kernel: Evision.Mat.t().

    structuring element used for dilation; if element=Mat(), a 3 x 3 rectangular structuring element is used. Kernel can be created using #getStructuringElement

Keyword Arguments
  • anchor: Point.

    position of the anchor within the element; default value (-1, -1) means that the anchor is at the element center.

  • iterations: int.

    number of times dilation is applied.

  • borderType: int.

    pixel extrapolation method, see #BorderTypes. #BORDER_WRAP is not suported.

  • borderValue: Scalar.

    border value in case of a constant border

Return
  • dst: Evision.Mat.t().

    output image of the same size and type as src.

The function dilates the source image using the specified structuring element that determines the shape of a pixel neighborhood over which the maximum is taken: \f[\texttt{dst} (x,y) = \max _{(x',y'): \, \texttt{element} (x',y') \ne0 } \texttt{src} (x+x',y+y')\f] The function supports the in-place mode. Dilation can be applied several ( iterations ) times. In case of multi-channel images, each channel is processed independently. @sa erode, morphologyEx, getStructuringElement

Python prototype (for reference only):

dilate(src, kernel[, dst[, anchor[, iterations[, borderType[, borderValue]]]]]) -> dst
Link to this function

dilate(src, kernel, opts)

View Source
@spec dilate(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Dilates an image by using a specific structuring element.

Positional Arguments
  • src: Evision.Mat.t().

    input image; the number of channels can be arbitrary, but the depth should be one of CV_8U, CV_16U, CV_16S, CV_32F or CV_64F.

  • kernel: Evision.Mat.t().

    structuring element used for dilation; if element=Mat(), a 3 x 3 rectangular structuring element is used. Kernel can be created using #getStructuringElement

Keyword Arguments
  • anchor: Point.

    position of the anchor within the element; default value (-1, -1) means that the anchor is at the element center.

  • iterations: int.

    number of times dilation is applied.

  • borderType: int.

    pixel extrapolation method, see #BorderTypes. #BORDER_WRAP is not suported.

  • borderValue: Scalar.

    border value in case of a constant border

Return
  • dst: Evision.Mat.t().

    output image of the same size and type as src.

The function dilates the source image using the specified structuring element that determines the shape of a pixel neighborhood over which the maximum is taken: \f[\texttt{dst} (x,y) = \max _{(x',y'): \, \texttt{element} (x',y') \ne0 } \texttt{src} (x+x',y+y')\f] The function supports the in-place mode. Dilation can be applied several ( iterations ) times. In case of multi-channel images, each channel is processed independently. @sa erode, morphologyEx, getStructuringElement

Python prototype (for reference only):

dilate(src, kernel[, dst[, anchor[, iterations[, borderType[, borderValue]]]]]) -> dst
Link to this function

displayOverlay(winname, text)

View Source
@spec displayOverlay(binary(), binary()) :: :ok | {:error, String.t()}

Displays a text on a window image as an overlay for a specified duration.

Positional Arguments
  • winname: String.

    Name of the window.

  • text: String.

    Overlay text to write on a window image.

Keyword Arguments
  • delayms: int.

    The period (in milliseconds), during which the overlay text is displayed. If this function is called before the previous overlay text timed out, the timer is restarted and the text is updated. If this value is zero, the text never disappears.

The function displayOverlay displays useful information/tips on top of the window for a certain amount of time delayms. The function does not modify the image, displayed in the window, that is, after the specified delay the original content of the window is restored.

Python prototype (for reference only):

displayOverlay(winname, text[, delayms]) -> None
Link to this function

displayOverlay(winname, text, opts)

View Source
@spec displayOverlay(binary(), binary(), [{atom(), term()}, ...] | nil) ::
  :ok | {:error, String.t()}

Displays a text on a window image as an overlay for a specified duration.

Positional Arguments
  • winname: String.

    Name of the window.

  • text: String.

    Overlay text to write on a window image.

Keyword Arguments
  • delayms: int.

    The period (in milliseconds), during which the overlay text is displayed. If this function is called before the previous overlay text timed out, the timer is restarted and the text is updated. If this value is zero, the text never disappears.

The function displayOverlay displays useful information/tips on top of the window for a certain amount of time delayms. The function does not modify the image, displayed in the window, that is, after the specified delay the original content of the window is restored.

Python prototype (for reference only):

displayOverlay(winname, text[, delayms]) -> None
Link to this function

displayStatusBar(winname, text)

View Source
@spec displayStatusBar(binary(), binary()) :: :ok | {:error, String.t()}

Displays a text on the window statusbar during the specified period of time.

Positional Arguments
  • winname: String.

    Name of the window.

  • text: String.

    Text to write on the window statusbar.

Keyword Arguments
  • delayms: int.

    Duration (in milliseconds) to display the text. If this function is called before the previous text timed out, the timer is restarted and the text is updated. If this value is zero, the text never disappears.

The function displayStatusBar displays useful information/tips on top of the window for a certain amount of time delayms . This information is displayed on the window statusbar (the window must be created with the CV_GUI_EXPANDED flags).

Python prototype (for reference only):

displayStatusBar(winname, text[, delayms]) -> None
Link to this function

displayStatusBar(winname, text, opts)

View Source
@spec displayStatusBar(binary(), binary(), [{atom(), term()}, ...] | nil) ::
  :ok | {:error, String.t()}

Displays a text on the window statusbar during the specified period of time.

Positional Arguments
  • winname: String.

    Name of the window.

  • text: String.

    Text to write on the window statusbar.

Keyword Arguments
  • delayms: int.

    Duration (in milliseconds) to display the text. If this function is called before the previous text timed out, the timer is restarted and the text is updated. If this value is zero, the text never disappears.

The function displayStatusBar displays useful information/tips on top of the window for a certain amount of time delayms . This information is displayed on the window statusbar (the window must be created with the CV_GUI_EXPANDED flags).

Python prototype (for reference only):

displayStatusBar(winname, text[, delayms]) -> None
Link to this function

distanceTransform(src, distanceType, maskSize)

View Source
@spec distanceTransform(Evision.Mat.maybe_mat_in(), integer(), integer()) ::
  Evision.Mat.t() | {:error, String.t()}

distanceTransform

Positional Arguments
  • src: Evision.Mat.t().

    8-bit, single-channel (binary) source image.

  • distanceType: int.

    Type of distance, see #DistanceTypes

  • maskSize: int.

    Size of the distance transform mask, see #DistanceTransformMasks. In case of the #DIST_L1 or #DIST_C distance type, the parameter is forced to 3 because a \f$3\times 3\f$ mask gives the same result as \f$5\times 5\f$ or any larger aperture.

Keyword Arguments
  • dstType: int.

    Type of output image. It can be CV_8U or CV_32F. Type CV_8U can be used only for the first variant of the function and distanceType == #DIST_L1.

Return
  • dst: Evision.Mat.t().

    Output image with calculated distances. It is a 8-bit or 32-bit floating-point, single-channel image of the same size as src .

Has overloading in C++

Python prototype (for reference only):

distanceTransform(src, distanceType, maskSize[, dst[, dstType]]) -> dst
Link to this function

distanceTransform(src, distanceType, maskSize, opts)

View Source
@spec distanceTransform(
  Evision.Mat.maybe_mat_in(),
  integer(),
  integer(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

distanceTransform

Positional Arguments
  • src: Evision.Mat.t().

    8-bit, single-channel (binary) source image.

  • distanceType: int.

    Type of distance, see #DistanceTypes

  • maskSize: int.

    Size of the distance transform mask, see #DistanceTransformMasks. In case of the #DIST_L1 or #DIST_C distance type, the parameter is forced to 3 because a \f$3\times 3\f$ mask gives the same result as \f$5\times 5\f$ or any larger aperture.

Keyword Arguments
  • dstType: int.

    Type of output image. It can be CV_8U or CV_32F. Type CV_8U can be used only for the first variant of the function and distanceType == #DIST_L1.

Return
  • dst: Evision.Mat.t().

    Output image with calculated distances. It is a 8-bit or 32-bit floating-point, single-channel image of the same size as src .

Has overloading in C++

Python prototype (for reference only):

distanceTransform(src, distanceType, maskSize[, dst[, dstType]]) -> dst
Link to this function

distanceTransformWithLabels(src, distanceType, maskSize)

View Source
@spec distanceTransformWithLabels(Evision.Mat.maybe_mat_in(), integer(), integer()) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Calculates the distance to the closest zero pixel for each pixel of the source image.

Positional Arguments
  • src: Evision.Mat.t().

    8-bit, single-channel (binary) source image.

  • distanceType: int.

    Type of distance, see #DistanceTypes

  • maskSize: int.

    Size of the distance transform mask, see #DistanceTransformMasks. #DIST_MASK_PRECISE is not supported by this variant. In case of the #DIST_L1 or #DIST_C distance type, the parameter is forced to 3 because a \f$3\times 3\f$ mask gives the same result as \f$5\times 5\f$ or any larger aperture.

Keyword Arguments
  • labelType: int.

    Type of the label array to build, see #DistanceTransformLabelTypes.

Return
  • dst: Evision.Mat.t().

    Output image with calculated distances. It is a 8-bit or 32-bit floating-point, single-channel image of the same size as src.

  • labels: Evision.Mat.t().

    Output 2D array of labels (the discrete Voronoi diagram). It has the type CV_32SC1 and the same size as src.

The function cv::distanceTransform calculates the approximate or precise distance from every binary image pixel to the nearest zero pixel. For zero image pixels, the distance will obviously be zero. When maskSize == #DIST_MASK_PRECISE and distanceType == #DIST_L2 , the function runs the algorithm described in @cite Felzenszwalb04 . This algorithm is parallelized with the TBB library. In other cases, the algorithm @cite Borgefors86 is used. This means that for a pixel the function finds the shortest path to the nearest zero pixel consisting of basic shifts: horizontal, vertical, diagonal, or knight's move (the latest is available for a \f$5\times 5\f$ mask). The overall distance is calculated as a sum of these basic distances. Since the distance function should be symmetric, all of the horizontal and vertical shifts must have the same cost (denoted as a ), all the diagonal shifts must have the same cost (denoted as b), and all knight's moves must have the same cost (denoted as c). For the #DIST_C and #DIST_L1 types, the distance is calculated precisely, whereas for #DIST_L2 (Euclidean distance) the distance can be calculated only with a relative error (a \f$5\times 5\f$ mask gives more accurate results). For a,b, and c, OpenCV uses the values suggested in the original paper:

  • DIST_L1: a = 1, b = 2
  • DIST_L2:
  • 3 x 3: a=0.955, b=1.3693
  • 5 x 5: a=1, b=1.4, c=2.1969
  • DIST_C: a = 1, b = 1

Typically, for a fast, coarse distance estimation #DIST_L2, a \f$3\times 3\f$ mask is used. For a more accurate distance estimation #DIST_L2, a \f$5\times 5\f$ mask or the precise algorithm is used. Note that both the precise and the approximate algorithms are linear on the number of pixels. This variant of the function does not only compute the minimum distance for each pixel \f$(x, y)\f$ but also identifies the nearest connected component consisting of zero pixels (labelType==#DIST_LABEL_CCOMP) or the nearest zero pixel (labelType==#DIST_LABEL_PIXEL). Index of the component/pixel is stored in labels(x, y). When labelType==#DIST_LABEL_CCOMP, the function automatically finds connected components of zero pixels in the input image and marks them with distinct labels. When labelType==#DIST_LABEL_PIXEL, the function scans through the input image and marks all the zero pixels with distinct labels. In this mode, the complexity is still linear. That is, the function provides a very fast way to compute the Voronoi diagram for a binary image. Currently, the second variant can use only the approximate distance transform algorithm, i.e. maskSize=#DIST_MASK_PRECISE is not supported yet.

Python prototype (for reference only):

distanceTransformWithLabels(src, distanceType, maskSize[, dst[, labels[, labelType]]]) -> dst, labels
Link to this function

distanceTransformWithLabels(src, distanceType, maskSize, opts)

View Source
@spec distanceTransformWithLabels(
  Evision.Mat.maybe_mat_in(),
  integer(),
  integer(),
  [{atom(), term()}, ...] | nil
) :: {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Calculates the distance to the closest zero pixel for each pixel of the source image.

Positional Arguments
  • src: Evision.Mat.t().

    8-bit, single-channel (binary) source image.

  • distanceType: int.

    Type of distance, see #DistanceTypes

  • maskSize: int.

    Size of the distance transform mask, see #DistanceTransformMasks. #DIST_MASK_PRECISE is not supported by this variant. In case of the #DIST_L1 or #DIST_C distance type, the parameter is forced to 3 because a \f$3\times 3\f$ mask gives the same result as \f$5\times 5\f$ or any larger aperture.

Keyword Arguments
  • labelType: int.

    Type of the label array to build, see #DistanceTransformLabelTypes.

Return
  • dst: Evision.Mat.t().

    Output image with calculated distances. It is a 8-bit or 32-bit floating-point, single-channel image of the same size as src.

  • labels: Evision.Mat.t().

    Output 2D array of labels (the discrete Voronoi diagram). It has the type CV_32SC1 and the same size as src.

The function cv::distanceTransform calculates the approximate or precise distance from every binary image pixel to the nearest zero pixel. For zero image pixels, the distance will obviously be zero. When maskSize == #DIST_MASK_PRECISE and distanceType == #DIST_L2 , the function runs the algorithm described in @cite Felzenszwalb04 . This algorithm is parallelized with the TBB library. In other cases, the algorithm @cite Borgefors86 is used. This means that for a pixel the function finds the shortest path to the nearest zero pixel consisting of basic shifts: horizontal, vertical, diagonal, or knight's move (the latest is available for a \f$5\times 5\f$ mask). The overall distance is calculated as a sum of these basic distances. Since the distance function should be symmetric, all of the horizontal and vertical shifts must have the same cost (denoted as a ), all the diagonal shifts must have the same cost (denoted as b), and all knight's moves must have the same cost (denoted as c). For the #DIST_C and #DIST_L1 types, the distance is calculated precisely, whereas for #DIST_L2 (Euclidean distance) the distance can be calculated only with a relative error (a \f$5\times 5\f$ mask gives more accurate results). For a,b, and c, OpenCV uses the values suggested in the original paper:

  • DIST_L1: a = 1, b = 2
  • DIST_L2:
  • 3 x 3: a=0.955, b=1.3693
  • 5 x 5: a=1, b=1.4, c=2.1969
  • DIST_C: a = 1, b = 1

Typically, for a fast, coarse distance estimation #DIST_L2, a \f$3\times 3\f$ mask is used. For a more accurate distance estimation #DIST_L2, a \f$5\times 5\f$ mask or the precise algorithm is used. Note that both the precise and the approximate algorithms are linear on the number of pixels. This variant of the function does not only compute the minimum distance for each pixel \f$(x, y)\f$ but also identifies the nearest connected component consisting of zero pixels (labelType==#DIST_LABEL_CCOMP) or the nearest zero pixel (labelType==#DIST_LABEL_PIXEL). Index of the component/pixel is stored in labels(x, y). When labelType==#DIST_LABEL_CCOMP, the function automatically finds connected components of zero pixels in the input image and marks them with distinct labels. When labelType==#DIST_LABEL_PIXEL, the function scans through the input image and marks all the zero pixels with distinct labels. In this mode, the complexity is still linear. That is, the function provides a very fast way to compute the Voronoi diagram for a binary image. Currently, the second variant can use only the approximate distance transform algorithm, i.e. maskSize=#DIST_MASK_PRECISE is not supported yet.

Python prototype (for reference only):

distanceTransformWithLabels(src, distanceType, maskSize[, dst[, labels[, labelType]]]) -> dst, labels
@spec divide(number(), Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}
@spec divide(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

Variant 1:

divide

Positional Arguments
  • scale: double
  • src2: Evision.Mat.t()
Keyword Arguments
  • dtype: int.
Return
  • dst: Evision.Mat.t().

Has overloading in C++

Python prototype (for reference only):

divide(scale, src2[, dst[, dtype]]) -> dst

Variant 2:

Performs per-element division of two arrays or a scalar by an array.

Positional Arguments
  • src1: Evision.Mat.t().

    first input array.

  • src2: Evision.Mat.t().

    second input array of the same size and type as src1.

Keyword Arguments
  • scale: double.

    scalar factor.

  • dtype: int.

    optional depth of the output array; if -1, dst will have depth src2.depth(), but in case of an array-by-array division, you can only pass -1 when src1.depth()==src2.depth().

Return
  • dst: Evision.Mat.t().

    output array of the same size and type as src2.

The function cv::divide divides one array by another: \f[\texttt{dst(I) = saturate(src1(I)*scale/src2(I))}\f] or a scalar by an array when there is no src1 : \f[\texttt{dst(I) = saturate(scale/src2(I))}\f] Different channels of multi-channel arrays are processed independently. For integer types when src2(I) is zero, dst(I) will also be zero. Note: In case of floating point data there is no special defined behavior for zero src2(I) values. Regular floating-point division is used. Expect correct IEEE-754 behaviour for floating-point data (with NaN, Inf result values). Note: Saturation is not applied when the output array has the depth CV_32S. You may even get result of an incorrect sign in the case of overflow. Note: (Python) Be careful to difference behaviour between src1/src2 are single number and they are tuple/array. divide(src,X) means divide(src,(X,X,X,X)). divide(src,(X,)) means divide(src,(X,0,0,0)). @sa multiply, add, subtract

Python prototype (for reference only):

divide(src1, src2[, dst[, scale[, dtype]]]) -> dst
Link to this function

divide(scale, src2, opts)

View Source
@spec divide(number(), Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}
@spec divide(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Variant 1:

divide

Positional Arguments
  • scale: double
  • src2: Evision.Mat.t()
Keyword Arguments
  • dtype: int.
Return
  • dst: Evision.Mat.t().

Has overloading in C++

Python prototype (for reference only):

divide(scale, src2[, dst[, dtype]]) -> dst

Variant 2:

Performs per-element division of two arrays or a scalar by an array.

Positional Arguments
  • src1: Evision.Mat.t().

    first input array.

  • src2: Evision.Mat.t().

    second input array of the same size and type as src1.

Keyword Arguments
  • scale: double.

    scalar factor.

  • dtype: int.

    optional depth of the output array; if -1, dst will have depth src2.depth(), but in case of an array-by-array division, you can only pass -1 when src1.depth()==src2.depth().

Return
  • dst: Evision.Mat.t().

    output array of the same size and type as src2.

The function cv::divide divides one array by another: \f[\texttt{dst(I) = saturate(src1(I)*scale/src2(I))}\f] or a scalar by an array when there is no src1 : \f[\texttt{dst(I) = saturate(scale/src2(I))}\f] Different channels of multi-channel arrays are processed independently. For integer types when src2(I) is zero, dst(I) will also be zero. Note: In case of floating point data there is no special defined behavior for zero src2(I) values. Regular floating-point division is used. Expect correct IEEE-754 behaviour for floating-point data (with NaN, Inf result values). Note: Saturation is not applied when the output array has the depth CV_32S. You may even get result of an incorrect sign in the case of overflow. Note: (Python) Be careful to difference behaviour between src1/src2 are single number and they are tuple/array. divide(src,X) means divide(src,(X,X,X,X)). divide(src,(X,)) means divide(src,(X,0,0,0)). @sa multiply, add, subtract

Python prototype (for reference only):

divide(src1, src2[, dst[, scale[, dtype]]]) -> dst
Link to this function

divSpectrums(a, b, flags)

View Source
@spec divSpectrums(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in(), integer()) ::
  Evision.Mat.t() | {:error, String.t()}

Performs the per-element division of the first Fourier spectrum by the second Fourier spectrum.

Positional Arguments
  • a: Evision.Mat.t().

    first input array.

  • b: Evision.Mat.t().

    second input array of the same size and type as src1 .

  • flags: int.

    operation flags; currently, the only supported flag is cv::DFT_ROWS, which indicates that each row of src1 and src2 is an independent 1D Fourier spectrum. If you do not want to use this flag, then simply add a 0 as value.

Keyword Arguments
  • conjB: bool.

    optional flag that conjugates the second input array before the multiplication (true) or not (false).

Return
  • c: Evision.Mat.t().

    output array of the same size and type as src1 .

The function cv::divSpectrums performs the per-element division of the first array by the second array. The arrays are CCS-packed or complex matrices that are results of a real or complex Fourier transform.

Python prototype (for reference only):

divSpectrums(a, b, flags[, c[, conjB]]) -> c
Link to this function

divSpectrums(a, b, flags, opts)

View Source
@spec divSpectrums(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  integer(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Performs the per-element division of the first Fourier spectrum by the second Fourier spectrum.

Positional Arguments
  • a: Evision.Mat.t().

    first input array.

  • b: Evision.Mat.t().

    second input array of the same size and type as src1 .

  • flags: int.

    operation flags; currently, the only supported flag is cv::DFT_ROWS, which indicates that each row of src1 and src2 is an independent 1D Fourier spectrum. If you do not want to use this flag, then simply add a 0 as value.

Keyword Arguments
  • conjB: bool.

    optional flag that conjugates the second input array before the multiplication (true) or not (false).

Return
  • c: Evision.Mat.t().

    output array of the same size and type as src1 .

The function cv::divSpectrums performs the per-element division of the first array by the second array. The arrays are CCS-packed or complex matrices that are results of a real or complex Fourier transform.

Python prototype (for reference only):

divSpectrums(a, b, flags[, c[, conjB]]) -> c
Link to this function

drawChessboardCorners(image, patternSize, corners, patternWasFound)

View Source
@spec drawChessboardCorners(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  Evision.Mat.maybe_mat_in(),
  boolean()
) :: Evision.Mat.t() | {:error, String.t()}

Renders the detected chessboard corners.

Positional Arguments
  • patternSize: Size.

    Number of inner corners per a chessboard row and column (patternSize = cv::Size(points_per_row,points_per_column)).

  • corners: Evision.Mat.t().

    Array of detected corners, the output of #findChessboardCorners.

  • patternWasFound: bool.

    Parameter indicating whether the complete board was found or not. The return value of #findChessboardCorners should be passed here.

Return
  • image: Evision.Mat.t().

    Destination image. It must be an 8-bit color image.

The function draws individual chessboard corners detected either as red circles if the board was not found, or as colored corners connected with lines if the board was found.

Python prototype (for reference only):

drawChessboardCorners(image, patternSize, corners, patternWasFound) -> image
Link to this function

drawContours(image, contours, contourIdx, color)

View Source
@spec drawContours(
  Evision.Mat.maybe_mat_in(),
  [Evision.Mat.maybe_mat_in()],
  integer(),
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()}
) :: Evision.Mat.t() | {:error, String.t()}

Draws contours outlines or filled contours.

Positional Arguments
  • contours: [Evision.Mat].

    All the input contours. Each contour is stored as a point vector.

  • contourIdx: int.

    Parameter indicating a contour to draw. If it is negative, all the contours are drawn.

  • color: Scalar.

    Color of the contours.

Keyword Arguments
  • thickness: int.

    Thickness of lines the contours are drawn with. If it is negative (for example, thickness=#FILLED ), the contour interiors are drawn.

  • lineType: int.

    Line connectivity. See #LineTypes

  • hierarchy: Evision.Mat.t().

    Optional information about hierarchy. It is only needed if you want to draw only some of the contours (see maxLevel ).

  • maxLevel: int.

    Maximal level for drawn contours. If it is 0, only the specified contour is drawn. If it is 1, the function draws the contour(s) and all the nested contours. If it is 2, the function draws the contours, all the nested contours, all the nested-to-nested contours, and so on. This parameter is only taken into account when there is hierarchy available.

  • offset: Point.

    Optional contour shift parameter. Shift all the drawn contours by the specified \f$\texttt{offset}=(dx,dy)\f$ .

Return
  • image: Evision.Mat.t().

    Destination image.

The function draws contour outlines in the image if \f$\texttt{thickness} \ge 0\f$ or fills the area bounded by the contours if \f$\texttt{thickness}<0\f$ . The example below shows how to retrieve connected components from the binary image and label them: : @include snippets/imgproc_drawContours.cpp Note: When thickness=#FILLED, the function is designed to handle connected components with holes correctly even when no hierarchy data is provided. This is done by analyzing all the outlines together using even-odd rule. This may give incorrect results if you have a joint collection of separately retrieved contours. In order to solve this problem, you need to call #drawContours separately for each sub-group of contours, or iterate over the collection using contourIdx parameter.

Python prototype (for reference only):

drawContours(image, contours, contourIdx, color[, thickness[, lineType[, hierarchy[, maxLevel[, offset]]]]]) -> image
Link to this function

drawContours(image, contours, contourIdx, color, opts)

View Source
@spec drawContours(
  Evision.Mat.maybe_mat_in(),
  [Evision.Mat.maybe_mat_in()],
  integer(),
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()},
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Draws contours outlines or filled contours.

Positional Arguments
  • contours: [Evision.Mat].

    All the input contours. Each contour is stored as a point vector.

  • contourIdx: int.

    Parameter indicating a contour to draw. If it is negative, all the contours are drawn.

  • color: Scalar.

    Color of the contours.

Keyword Arguments
  • thickness: int.

    Thickness of lines the contours are drawn with. If it is negative (for example, thickness=#FILLED ), the contour interiors are drawn.

  • lineType: int.

    Line connectivity. See #LineTypes

  • hierarchy: Evision.Mat.t().

    Optional information about hierarchy. It is only needed if you want to draw only some of the contours (see maxLevel ).

  • maxLevel: int.

    Maximal level for drawn contours. If it is 0, only the specified contour is drawn. If it is 1, the function draws the contour(s) and all the nested contours. If it is 2, the function draws the contours, all the nested contours, all the nested-to-nested contours, and so on. This parameter is only taken into account when there is hierarchy available.

  • offset: Point.

    Optional contour shift parameter. Shift all the drawn contours by the specified \f$\texttt{offset}=(dx,dy)\f$ .

Return
  • image: Evision.Mat.t().

    Destination image.

The function draws contour outlines in the image if \f$\texttt{thickness} \ge 0\f$ or fills the area bounded by the contours if \f$\texttt{thickness}<0\f$ . The example below shows how to retrieve connected components from the binary image and label them: : @include snippets/imgproc_drawContours.cpp Note: When thickness=#FILLED, the function is designed to handle connected components with holes correctly even when no hierarchy data is provided. This is done by analyzing all the outlines together using even-odd rule. This may give incorrect results if you have a joint collection of separately retrieved contours. In order to solve this problem, you need to call #drawContours separately for each sub-group of contours, or iterate over the collection using contourIdx parameter.

Python prototype (for reference only):

drawContours(image, contours, contourIdx, color[, thickness[, lineType[, hierarchy[, maxLevel[, offset]]]]]) -> image
Link to this function

drawFrameAxes(image, cameraMatrix, distCoeffs, rvec, tvec, length)

View Source

Draw axes of the world/object coordinate system from pose estimation. @sa solvePnP

Positional Arguments
  • cameraMatrix: Evision.Mat.t().

    Input 3x3 floating-point matrix of camera intrinsic parameters. \f$\cameramatrix{A}\f$

  • distCoeffs: Evision.Mat.t().

    Input vector of distortion coefficients \f$\distcoeffs\f$. If the vector is empty, the zero distortion coefficients are assumed.

  • rvec: Evision.Mat.t().

    Rotation vector (see @ref Rodrigues ) that, together with tvec, brings points from the model coordinate system to the camera coordinate system.

  • tvec: Evision.Mat.t().

    Translation vector.

  • length: float.

    Length of the painted axes in the same unit than tvec (usually in meters).

Keyword Arguments
  • thickness: int.

    Line thickness of the painted axes.

Return
  • image: Evision.Mat.t().

    Input/output image. It must have 1 or 3 channels. The number of channels is not altered.

This function draws the axes of the world/object coordinate system w.r.t. to the camera frame. OX is drawn in red, OY in green and OZ in blue.

Python prototype (for reference only):

drawFrameAxes(image, cameraMatrix, distCoeffs, rvec, tvec, length[, thickness]) -> image
Link to this function

drawFrameAxes(image, cameraMatrix, distCoeffs, rvec, tvec, length, opts)

View Source

Draw axes of the world/object coordinate system from pose estimation. @sa solvePnP

Positional Arguments
  • cameraMatrix: Evision.Mat.t().

    Input 3x3 floating-point matrix of camera intrinsic parameters. \f$\cameramatrix{A}\f$

  • distCoeffs: Evision.Mat.t().

    Input vector of distortion coefficients \f$\distcoeffs\f$. If the vector is empty, the zero distortion coefficients are assumed.

  • rvec: Evision.Mat.t().

    Rotation vector (see @ref Rodrigues ) that, together with tvec, brings points from the model coordinate system to the camera coordinate system.

  • tvec: Evision.Mat.t().

    Translation vector.

  • length: float.

    Length of the painted axes in the same unit than tvec (usually in meters).

Keyword Arguments
  • thickness: int.

    Line thickness of the painted axes.

Return
  • image: Evision.Mat.t().

    Input/output image. It must have 1 or 3 channels. The number of channels is not altered.

This function draws the axes of the world/object coordinate system w.r.t. to the camera frame. OX is drawn in red, OY in green and OZ in blue.

Python prototype (for reference only):

drawFrameAxes(image, cameraMatrix, distCoeffs, rvec, tvec, length[, thickness]) -> image
Link to this function

drawKeypoints(image, keypoints, outImage)

View Source
@spec drawKeypoints(
  Evision.Mat.maybe_mat_in(),
  [Evision.KeyPoint.t()],
  Evision.Mat.maybe_mat_in()
) ::
  Evision.Mat.t() | {:error, String.t()}

Draws keypoints.

Positional Arguments
  • image: Evision.Mat.t().

    Source image.

  • keypoints: [Evision.KeyPoint].

    Keypoints from the source image.

Keyword Arguments
  • color: Scalar.

    Color of keypoints.

  • flags: DrawMatchesFlags.

    Flags setting drawing features. Possible flags bit values are defined by DrawMatchesFlags. See details above in drawMatches .

Return
  • outImage: Evision.Mat.t().

    Output image. Its content depends on the flags value defining what is drawn in the output image. See possible flags bit values below.

Note: For Python API, flags are modified as cv.DRAW_MATCHES_FLAGS_DEFAULT, cv.DRAW_MATCHES_FLAGS_DRAW_RICH_KEYPOINTS, cv.DRAW_MATCHES_FLAGS_DRAW_OVER_OUTIMG, cv.DRAW_MATCHES_FLAGS_NOT_DRAW_SINGLE_POINTS

Python prototype (for reference only):

drawKeypoints(image, keypoints, outImage[, color[, flags]]) -> outImage
Link to this function

drawKeypoints(image, keypoints, outImage, opts)

View Source
@spec drawKeypoints(
  Evision.Mat.maybe_mat_in(),
  [Evision.KeyPoint.t()],
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Draws keypoints.

Positional Arguments
  • image: Evision.Mat.t().

    Source image.

  • keypoints: [Evision.KeyPoint].

    Keypoints from the source image.

Keyword Arguments
  • color: Scalar.

    Color of keypoints.

  • flags: DrawMatchesFlags.

    Flags setting drawing features. Possible flags bit values are defined by DrawMatchesFlags. See details above in drawMatches .

Return
  • outImage: Evision.Mat.t().

    Output image. Its content depends on the flags value defining what is drawn in the output image. See possible flags bit values below.

Note: For Python API, flags are modified as cv.DRAW_MATCHES_FLAGS_DEFAULT, cv.DRAW_MATCHES_FLAGS_DRAW_RICH_KEYPOINTS, cv.DRAW_MATCHES_FLAGS_DRAW_OVER_OUTIMG, cv.DRAW_MATCHES_FLAGS_NOT_DRAW_SINGLE_POINTS

Python prototype (for reference only):

drawKeypoints(image, keypoints, outImage[, color[, flags]]) -> outImage
Link to this function

drawMarker(img, position, color)

View Source
@spec drawMarker(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()}
) :: Evision.Mat.t() | {:error, String.t()}

Draws a marker on a predefined position in an image.

Positional Arguments
  • position: Point.

    The point where the crosshair is positioned.

  • color: Scalar.

    Line color.

Keyword Arguments
  • markerType: int.

    The specific type of marker you want to use, see #MarkerTypes

  • markerSize: int.

    The length of the marker axis [default = 20 pixels]

  • thickness: int.

    Line thickness.

  • line_type: int.

    Type of the line, See #LineTypes

Return
  • img: Evision.Mat.t().

    Image.

The function cv::drawMarker draws a marker on a given position in the image. For the moment several marker types are supported, see #MarkerTypes for more information.

Python prototype (for reference only):

drawMarker(img, position, color[, markerType[, markerSize[, thickness[, line_type]]]]) -> img
Link to this function

drawMarker(img, position, color, opts)

View Source
@spec drawMarker(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()},
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Draws a marker on a predefined position in an image.

Positional Arguments
  • position: Point.

    The point where the crosshair is positioned.

  • color: Scalar.

    Line color.

Keyword Arguments
  • markerType: int.

    The specific type of marker you want to use, see #MarkerTypes

  • markerSize: int.

    The length of the marker axis [default = 20 pixels]

  • thickness: int.

    Line thickness.

  • line_type: int.

    Type of the line, See #LineTypes

Return
  • img: Evision.Mat.t().

    Image.

The function cv::drawMarker draws a marker on a given position in the image. For the moment several marker types are supported, see #MarkerTypes for more information.

Python prototype (for reference only):

drawMarker(img, position, color[, markerType[, markerSize[, thickness[, line_type]]]]) -> img
Link to this function

drawMatches(img1, keypoints1, img2, keypoints2, matches1to2, outImg)

View Source

Draws the found matches of keypoints from two images.

Positional Arguments
  • img1: Evision.Mat.t().

    First source image.

  • keypoints1: [Evision.KeyPoint].

    Keypoints from the first source image.

  • img2: Evision.Mat.t().

    Second source image.

  • keypoints2: [Evision.KeyPoint].

    Keypoints from the second source image.

  • matches1to2: [Evision.DMatch].

    Matches from the first image to the second one, which means that keypoints1[i] has a corresponding point in keypoints2[matches[i]] .

Keyword Arguments
  • matchColor: Scalar.

    Color of matches (lines and connected keypoints). If matchColor==Scalar::all(-1) , the color is generated randomly.

  • singlePointColor: Scalar.

    Color of single keypoints (circles), which means that keypoints do not have the matches. If singlePointColor==Scalar::all(-1) , the color is generated randomly.

  • matchesMask: [char].

    Mask determining which matches are drawn. If the mask is empty, all matches are drawn.

  • flags: DrawMatchesFlags.

    Flags setting drawing features. Possible flags bit values are defined by DrawMatchesFlags.

Return
  • outImg: Evision.Mat.t().

    Output image. Its content depends on the flags value defining what is drawn in the output image. See possible flags bit values below.

This function draws matches of keypoints from two images in the output image. Match is a line connecting two keypoints (circles). See cv::DrawMatchesFlags.

Python prototype (for reference only):

drawMatches(img1, keypoints1, img2, keypoints2, matches1to2, outImg[, matchColor[, singlePointColor[, matchesMask[, flags]]]]) -> outImg
Link to this function

drawMatches(img1, keypoints1, img2, keypoints2, matches1to2, outImg, opts)

View Source

Variant 1:

drawMatches

Positional Arguments
  • img1: Evision.Mat.t()
  • keypoints1: [Evision.KeyPoint]
  • img2: Evision.Mat.t()
  • keypoints2: [Evision.KeyPoint]
  • matches1to2: [Evision.DMatch]
  • matchesThickness: int
Keyword Arguments
  • matchColor: Scalar.
  • singlePointColor: Scalar.
  • matchesMask: [char].
  • flags: DrawMatchesFlags.
Return
  • outImg: Evision.Mat.t()

Has overloading in C++

Python prototype (for reference only):

drawMatches(img1, keypoints1, img2, keypoints2, matches1to2, outImg, matchesThickness[, matchColor[, singlePointColor[, matchesMask[, flags]]]]) -> outImg

Variant 2:

Draws the found matches of keypoints from two images.

Positional Arguments
  • img1: Evision.Mat.t().

    First source image.

  • keypoints1: [Evision.KeyPoint].

    Keypoints from the first source image.

  • img2: Evision.Mat.t().

    Second source image.

  • keypoints2: [Evision.KeyPoint].

    Keypoints from the second source image.

  • matches1to2: [Evision.DMatch].

    Matches from the first image to the second one, which means that keypoints1[i] has a corresponding point in keypoints2[matches[i]] .

Keyword Arguments
  • matchColor: Scalar.

    Color of matches (lines and connected keypoints). If matchColor==Scalar::all(-1) , the color is generated randomly.

  • singlePointColor: Scalar.

    Color of single keypoints (circles), which means that keypoints do not have the matches. If singlePointColor==Scalar::all(-1) , the color is generated randomly.

  • matchesMask: [char].

    Mask determining which matches are drawn. If the mask is empty, all matches are drawn.

  • flags: DrawMatchesFlags.

    Flags setting drawing features. Possible flags bit values are defined by DrawMatchesFlags.

Return
  • outImg: Evision.Mat.t().

    Output image. Its content depends on the flags value defining what is drawn in the output image. See possible flags bit values below.

This function draws matches of keypoints from two images in the output image. Match is a line connecting two keypoints (circles). See cv::DrawMatchesFlags.

Python prototype (for reference only):

drawMatches(img1, keypoints1, img2, keypoints2, matches1to2, outImg[, matchColor[, singlePointColor[, matchesMask[, flags]]]]) -> outImg
Link to this function

drawMatches(img1, keypoints1, img2, keypoints2, matches1to2, outImg, matchesThickness, opts)

View Source

drawMatches

Positional Arguments
  • img1: Evision.Mat.t()
  • keypoints1: [Evision.KeyPoint]
  • img2: Evision.Mat.t()
  • keypoints2: [Evision.KeyPoint]
  • matches1to2: [Evision.DMatch]
  • matchesThickness: int
Keyword Arguments
  • matchColor: Scalar.
  • singlePointColor: Scalar.
  • matchesMask: [char].
  • flags: DrawMatchesFlags.
Return
  • outImg: Evision.Mat.t()

Has overloading in C++

Python prototype (for reference only):

drawMatches(img1, keypoints1, img2, keypoints2, matches1to2, outImg, matchesThickness[, matchColor[, singlePointColor[, matchesMask[, flags]]]]) -> outImg
Link to this function

drawMatchesKnn(img1, keypoints1, img2, keypoints2, matches1to2, outImg)

View Source

drawMatchesKnn

Positional Arguments
  • img1: Evision.Mat.t()
  • keypoints1: [Evision.KeyPoint]
  • img2: Evision.Mat.t()
  • keypoints2: [Evision.KeyPoint]
  • matches1to2: [[Evision.DMatch]]
Keyword Arguments
  • matchColor: Scalar.
  • singlePointColor: Scalar.
  • matchesMask: [[char]].
  • flags: DrawMatchesFlags.
Return
  • outImg: Evision.Mat.t()

Python prototype (for reference only):

drawMatchesKnn(img1, keypoints1, img2, keypoints2, matches1to2, outImg[, matchColor[, singlePointColor[, matchesMask[, flags]]]]) -> outImg
Link to this function

drawMatchesKnn(img1, keypoints1, img2, keypoints2, matches1to2, outImg, opts)

View Source

drawMatchesKnn

Positional Arguments
  • img1: Evision.Mat.t()
  • keypoints1: [Evision.KeyPoint]
  • img2: Evision.Mat.t()
  • keypoints2: [Evision.KeyPoint]
  • matches1to2: [[Evision.DMatch]]
Keyword Arguments
  • matchColor: Scalar.
  • singlePointColor: Scalar.
  • matchesMask: [[char]].
  • flags: DrawMatchesFlags.
Return
  • outImg: Evision.Mat.t()

Python prototype (for reference only):

drawMatchesKnn(img1, keypoints1, img2, keypoints2, matches1to2, outImg[, matchColor[, singlePointColor[, matchesMask[, flags]]]]) -> outImg
Link to this function

edgePreservingFilter(src)

View Source
@spec edgePreservingFilter(Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

Filtering is the fundamental operation in image and video processing. Edge-preserving smoothing filters are used in many different applications @cite EM11 .

Positional Arguments
  • src: Evision.Mat.t().

    Input 8-bit 3-channel image.

Keyword Arguments
  • flags: int.

    Edge preserving filters: cv::RECURS_FILTER or cv::NORMCONV_FILTER

  • sigma_s: float.

    %Range between 0 to 200.

  • sigma_r: float.

    %Range between 0 to 1.

Return
  • dst: Evision.Mat.t().

    Output 8-bit 3-channel image.

Python prototype (for reference only):

edgePreservingFilter(src[, dst[, flags[, sigma_s[, sigma_r]]]]) -> dst
Link to this function

edgePreservingFilter(src, opts)

View Source
@spec edgePreservingFilter(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Filtering is the fundamental operation in image and video processing. Edge-preserving smoothing filters are used in many different applications @cite EM11 .

Positional Arguments
  • src: Evision.Mat.t().

    Input 8-bit 3-channel image.

Keyword Arguments
  • flags: int.

    Edge preserving filters: cv::RECURS_FILTER or cv::NORMCONV_FILTER

  • sigma_s: float.

    %Range between 0 to 200.

  • sigma_r: float.

    %Range between 0 to 1.

Return
  • dst: Evision.Mat.t().

    Output 8-bit 3-channel image.

Python prototype (for reference only):

edgePreservingFilter(src[, dst[, flags[, sigma_s[, sigma_r]]]]) -> dst
@spec eigen(Evision.Mat.maybe_mat_in()) ::
  {Evision.Mat.t(), Evision.Mat.t()} | false | {:error, String.t()}

Calculates eigenvalues and eigenvectors of a symmetric matrix.

Positional Arguments
  • src: Evision.Mat.t().

    input matrix that must have CV_32FC1 or CV_64FC1 type, square size and be symmetrical (src ^T^ == src).

Return
  • retval: bool

  • eigenvalues: Evision.Mat.t().

    output vector of eigenvalues of the same type as src; the eigenvalues are stored in the descending order.

  • eigenvectors: Evision.Mat.t().

    output matrix of eigenvectors; it has the same size and type as src; the eigenvectors are stored as subsequent matrix rows, in the same order as the corresponding eigenvalues.

The function cv::eigen calculates just eigenvalues, or eigenvalues and eigenvectors of the symmetric matrix src:

src*eigenvectors.row(i).t() = eigenvalues.at<srcType>(i)*eigenvectors.row(i).t()

Note: Use cv::eigenNonSymmetric for calculation of real eigenvalues and eigenvectors of non-symmetric matrix. @sa eigenNonSymmetric, completeSymm , PCA

Python prototype (for reference only):

eigen(src[, eigenvalues[, eigenvectors]]) -> retval, eigenvalues, eigenvectors
@spec eigen(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  {Evision.Mat.t(), Evision.Mat.t()} | false | {:error, String.t()}

Calculates eigenvalues and eigenvectors of a symmetric matrix.

Positional Arguments
  • src: Evision.Mat.t().

    input matrix that must have CV_32FC1 or CV_64FC1 type, square size and be symmetrical (src ^T^ == src).

Return
  • retval: bool

  • eigenvalues: Evision.Mat.t().

    output vector of eigenvalues of the same type as src; the eigenvalues are stored in the descending order.

  • eigenvectors: Evision.Mat.t().

    output matrix of eigenvectors; it has the same size and type as src; the eigenvectors are stored as subsequent matrix rows, in the same order as the corresponding eigenvalues.

The function cv::eigen calculates just eigenvalues, or eigenvalues and eigenvectors of the symmetric matrix src:

src*eigenvectors.row(i).t() = eigenvalues.at<srcType>(i)*eigenvectors.row(i).t()

Note: Use cv::eigenNonSymmetric for calculation of real eigenvalues and eigenvectors of non-symmetric matrix. @sa eigenNonSymmetric, completeSymm , PCA

Python prototype (for reference only):

eigen(src[, eigenvalues[, eigenvectors]]) -> retval, eigenvalues, eigenvectors
@spec eigenNonSymmetric(Evision.Mat.maybe_mat_in()) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Calculates eigenvalues and eigenvectors of a non-symmetric matrix (real eigenvalues only).

Positional Arguments
  • src: Evision.Mat.t().

    input matrix (CV_32FC1 or CV_64FC1 type).

Return
  • eigenvalues: Evision.Mat.t().

    output vector of eigenvalues (type is the same type as src).

  • eigenvectors: Evision.Mat.t().

    output matrix of eigenvectors (type is the same type as src). The eigenvectors are stored as subsequent matrix rows, in the same order as the corresponding eigenvalues.

Note: Assumes real eigenvalues. The function calculates eigenvalues and eigenvectors (optional) of the square matrix src:

src*eigenvectors.row(i).t() = eigenvalues.at<srcType>(i)*eigenvectors.row(i).t()

@sa eigen

Python prototype (for reference only):

eigenNonSymmetric(src[, eigenvalues[, eigenvectors]]) -> eigenvalues, eigenvectors
Link to this function

eigenNonSymmetric(src, opts)

View Source
@spec eigenNonSymmetric(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Calculates eigenvalues and eigenvectors of a non-symmetric matrix (real eigenvalues only).

Positional Arguments
  • src: Evision.Mat.t().

    input matrix (CV_32FC1 or CV_64FC1 type).

Return
  • eigenvalues: Evision.Mat.t().

    output vector of eigenvalues (type is the same type as src).

  • eigenvectors: Evision.Mat.t().

    output matrix of eigenvectors (type is the same type as src). The eigenvectors are stored as subsequent matrix rows, in the same order as the corresponding eigenvalues.

Note: Assumes real eigenvalues. The function calculates eigenvalues and eigenvectors (optional) of the square matrix src:

src*eigenvectors.row(i).t() = eigenvalues.at<srcType>(i)*eigenvectors.row(i).t()

@sa eigen

Python prototype (for reference only):

eigenNonSymmetric(src[, eigenvalues[, eigenvectors]]) -> eigenvalues, eigenvectors
Link to this function

ellipse2Poly(center, axes, angle, arcStart, arcEnd, delta)

View Source
@spec ellipse2Poly(
  {number(), number()},
  {number(), number()},
  integer(),
  integer(),
  integer(),
  integer()
) :: [{number(), number()}] | {:error, String.t()}

Approximates an elliptic arc with a polyline.

Positional Arguments
  • center: Point.

    Center of the arc.

  • axes: Size.

    Half of the size of the ellipse main axes. See #ellipse for details.

  • angle: int.

    Rotation angle of the ellipse in degrees. See #ellipse for details.

  • arcStart: int.

    Starting angle of the elliptic arc in degrees.

  • arcEnd: int.

    Ending angle of the elliptic arc in degrees.

  • delta: int.

    Angle between the subsequent polyline vertices. It defines the approximation accuracy.

Return
  • pts: [Point].

    Output vector of polyline vertices.

The function ellipse2Poly computes the vertices of a polyline that approximates the specified elliptic arc. It is used by #ellipse. If arcStart is greater than arcEnd, they are swapped.

Python prototype (for reference only):

ellipse2Poly(center, axes, angle, arcStart, arcEnd, delta) -> pts
Link to this function

ellipse(img, box, color)

View Source
@spec ellipse(
  Evision.Mat.maybe_mat_in(),
  Evision.RotatedRect.t(),
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()}
) :: Evision.Mat.t() | {:error, String.t()}

ellipse

Positional Arguments
  • box: {centre={x, y}, size={s1, s2}, angle}.

    Alternative ellipse representation via RotatedRect. This means that the function draws an ellipse inscribed in the rotated rectangle.

  • color: Scalar.

    Ellipse color.

Keyword Arguments
  • thickness: int.

    Thickness of the ellipse arc outline, if positive. Otherwise, this indicates that a filled ellipse sector is to be drawn.

  • lineType: int.

    Type of the ellipse boundary. See #LineTypes

Return
  • img: Evision.Mat.t().

    Image.

Has overloading in C++

Python prototype (for reference only):

ellipse(img, box, color[, thickness[, lineType]]) -> img
Link to this function

ellipse(img, box, color, opts)

View Source
@spec ellipse(
  Evision.Mat.maybe_mat_in(),
  Evision.RotatedRect.t(),
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()},
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

ellipse

Positional Arguments
  • box: {centre={x, y}, size={s1, s2}, angle}.

    Alternative ellipse representation via RotatedRect. This means that the function draws an ellipse inscribed in the rotated rectangle.

  • color: Scalar.

    Ellipse color.

Keyword Arguments
  • thickness: int.

    Thickness of the ellipse arc outline, if positive. Otherwise, this indicates that a filled ellipse sector is to be drawn.

  • lineType: int.

    Type of the ellipse boundary. See #LineTypes

Return
  • img: Evision.Mat.t().

    Image.

Has overloading in C++

Python prototype (for reference only):

ellipse(img, box, color[, thickness[, lineType]]) -> img
Link to this function

ellipse(img, center, axes, angle, startAngle, endAngle, color)

View Source
@spec ellipse(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  {number(), number()},
  number(),
  number(),
  number(),
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()}
) :: Evision.Mat.t() | {:error, String.t()}

Draws a simple or thick elliptic arc or fills an ellipse sector.

Positional Arguments
  • center: Point.

    Center of the ellipse.

  • axes: Size.

    Half of the size of the ellipse main axes.

  • angle: double.

    Ellipse rotation angle in degrees.

  • startAngle: double.

    Starting angle of the elliptic arc in degrees.

  • endAngle: double.

    Ending angle of the elliptic arc in degrees.

  • color: Scalar.

    Ellipse color.

Keyword Arguments
  • thickness: int.

    Thickness of the ellipse arc outline, if positive. Otherwise, this indicates that a filled ellipse sector is to be drawn.

  • lineType: int.

    Type of the ellipse boundary. See #LineTypes

  • shift: int.

    Number of fractional bits in the coordinates of the center and values of axes.

Return
  • img: Evision.Mat.t().

    Image.

The function cv::ellipse with more parameters draws an ellipse outline, a filled ellipse, an elliptic arc, or a filled ellipse sector. The drawing code uses general parametric form. A piecewise-linear curve is used to approximate the elliptic arc boundary. If you need more control of the ellipse rendering, you can retrieve the curve using #ellipse2Poly and then render it with #polylines or fill it with #fillPoly. If you use the first variant of the function and want to draw the whole ellipse, not an arc, pass startAngle=0 and endAngle=360. If startAngle is greater than endAngle, they are swapped. The figure below explains the meaning of the parameters to draw the blue arc. Parameters of Elliptic Arc

Python prototype (for reference only):

ellipse(img, center, axes, angle, startAngle, endAngle, color[, thickness[, lineType[, shift]]]) -> img
Link to this function

ellipse(img, center, axes, angle, startAngle, endAngle, color, opts)

View Source
@spec ellipse(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  {number(), number()},
  number(),
  number(),
  number(),
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()},
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Draws a simple or thick elliptic arc or fills an ellipse sector.

Positional Arguments
  • center: Point.

    Center of the ellipse.

  • axes: Size.

    Half of the size of the ellipse main axes.

  • angle: double.

    Ellipse rotation angle in degrees.

  • startAngle: double.

    Starting angle of the elliptic arc in degrees.

  • endAngle: double.

    Ending angle of the elliptic arc in degrees.

  • color: Scalar.

    Ellipse color.

Keyword Arguments
  • thickness: int.

    Thickness of the ellipse arc outline, if positive. Otherwise, this indicates that a filled ellipse sector is to be drawn.

  • lineType: int.

    Type of the ellipse boundary. See #LineTypes

  • shift: int.

    Number of fractional bits in the coordinates of the center and values of axes.

Return
  • img: Evision.Mat.t().

    Image.

The function cv::ellipse with more parameters draws an ellipse outline, a filled ellipse, an elliptic arc, or a filled ellipse sector. The drawing code uses general parametric form. A piecewise-linear curve is used to approximate the elliptic arc boundary. If you need more control of the ellipse rendering, you can retrieve the curve using #ellipse2Poly and then render it with #polylines or fill it with #fillPoly. If you use the first variant of the function and want to draw the whole ellipse, not an arc, pass startAngle=0 and endAngle=360. If startAngle is greater than endAngle, they are swapped. The figure below explains the meaning of the parameters to draw the blue arc. Parameters of Elliptic Arc

Python prototype (for reference only):

ellipse(img, center, axes, angle, startAngle, endAngle, color[, thickness[, lineType[, shift]]]) -> img
Link to this function

emd(signature1, signature2, distType)

View Source

Computes the "minimal work" distance between two weighted point configurations.

Positional Arguments
  • signature1: Evision.Mat.t().

    First signature, a \f$\texttt{size1}\times \texttt{dims}+1\f$ floating-point matrix. Each row stores the point weight followed by the point coordinates. The matrix is allowed to have a single column (weights only) if the user-defined cost matrix is used. The weights must be non-negative and have at least one non-zero value.

  • signature2: Evision.Mat.t().

    Second signature of the same format as signature1 , though the number of rows may be different. The total weights may be different. In this case an extra "dummy" point is added to either signature1 or signature2. The weights must be non-negative and have at least one non-zero value.

  • distType: int.

    Used metric. See #DistanceTypes.

Keyword Arguments
  • cost: Evision.Mat.t().

    User-defined \f$\texttt{size1}\times \texttt{size2}\f$ cost matrix. Also, if a cost matrix is used, lower boundary lowerBound cannot be calculated because it needs a metric function.

Return
  • retval: float

  • lowerBound: float.

    Optional input/output parameter: lower boundary of a distance between the two signatures that is a distance between mass centers. The lower boundary may not be calculated if the user-defined cost matrix is used, the total weights of point configurations are not equal, or if the signatures consist of weights only (the signature matrices have a single column). You

    • must* initialize \lowerBound . If the calculated distance between mass centers is greater or equal to *lowerBound (it means that the signatures are far enough), the function does not calculate EMD. In any case *lowerBound is set to the calculated distance between mass centers on return. Thus, if you want to calculate both distance between mass centers and EMD, *lowerBound should be set to 0.
  • flow: Evision.Mat.t().

    Resultant \f$\texttt{size1} \times \texttt{size2}\f$ flow matrix: \f$\texttt{flow}_{i,j}\f$ is a flow from \f$i\f$ -th point of signature1 to \f$j\f$ -th point of signature2 .

The function computes the earth mover distance and/or a lower boundary of the distance between the two weighted point configurations. One of the applications described in @cite RubnerSept98, @cite Rubner2000 is multi-dimensional histogram comparison for image retrieval. EMD is a transportation problem that is solved using some modification of a simplex algorithm, thus the complexity is exponential in the worst case, though, on average it is much faster. In the case of a real metric the lower boundary can be calculated even faster (using linear-time algorithm) and it can be used to determine roughly whether the two signatures are far enough so that they cannot relate to the same object.

Python prototype (for reference only):

EMD(signature1, signature2, distType[, cost[, lowerBound[, flow]]]) -> retval, lowerBound, flow
Link to this function

emd(signature1, signature2, distType, opts)

View Source
@spec emd(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  integer(),
  [{atom(), term()}, ...] | nil
) :: {number(), number(), Evision.Mat.t()} | {:error, String.t()}

Computes the "minimal work" distance between two weighted point configurations.

Positional Arguments
  • signature1: Evision.Mat.t().

    First signature, a \f$\texttt{size1}\times \texttt{dims}+1\f$ floating-point matrix. Each row stores the point weight followed by the point coordinates. The matrix is allowed to have a single column (weights only) if the user-defined cost matrix is used. The weights must be non-negative and have at least one non-zero value.

  • signature2: Evision.Mat.t().

    Second signature of the same format as signature1 , though the number of rows may be different. The total weights may be different. In this case an extra "dummy" point is added to either signature1 or signature2. The weights must be non-negative and have at least one non-zero value.

  • distType: int.

    Used metric. See #DistanceTypes.

Keyword Arguments
  • cost: Evision.Mat.t().

    User-defined \f$\texttt{size1}\times \texttt{size2}\f$ cost matrix. Also, if a cost matrix is used, lower boundary lowerBound cannot be calculated because it needs a metric function.

Return
  • retval: float

  • lowerBound: float.

    Optional input/output parameter: lower boundary of a distance between the two signatures that is a distance between mass centers. The lower boundary may not be calculated if the user-defined cost matrix is used, the total weights of point configurations are not equal, or if the signatures consist of weights only (the signature matrices have a single column). You

    • must* initialize \lowerBound . If the calculated distance between mass centers is greater or equal to *lowerBound (it means that the signatures are far enough), the function does not calculate EMD. In any case *lowerBound is set to the calculated distance between mass centers on return. Thus, if you want to calculate both distance between mass centers and EMD, *lowerBound should be set to 0.
  • flow: Evision.Mat.t().

    Resultant \f$\texttt{size1} \times \texttt{size2}\f$ flow matrix: \f$\texttt{flow}_{i,j}\f$ is a flow from \f$i\f$ -th point of signature1 to \f$j\f$ -th point of signature2 .

The function computes the earth mover distance and/or a lower boundary of the distance between the two weighted point configurations. One of the applications described in @cite RubnerSept98, @cite Rubner2000 is multi-dimensional histogram comparison for image retrieval. EMD is a transportation problem that is solved using some modification of a simplex algorithm, thus the complexity is exponential in the worst case, though, on average it is much faster. In the case of a real metric the lower boundary can be calculated even faster (using linear-time algorithm) and it can be used to determine roughly whether the two signatures are far enough so that they cannot relate to the same object.

Python prototype (for reference only):

EMD(signature1, signature2, distType[, cost[, lowerBound[, flow]]]) -> retval, lowerBound, flow

return a list of enabled modules in this build

@spec equalizeHist(Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

Equalizes the histogram of a grayscale image.

Positional Arguments
  • src: Evision.Mat.t().

    Source 8-bit single channel image.

Return
  • dst: Evision.Mat.t().

    Destination image of the same size and type as src .

The function equalizes the histogram of the input image using the following algorithm:

  • Calculate the histogram \f$H\f$ for src .

  • Normalize the histogram so that the sum of histogram bins is 255.

  • Compute the integral of the histogram: \f[H'_i = \sum _{0 \le j < i} H(j)\f]

  • Transform the image using \f$H'\f$ as a look-up table: \f$\texttt{dst}(x,y) = H'(\texttt{src}(x,y))\f$

The algorithm normalizes the brightness and increases the contrast of the image.

Python prototype (for reference only):

equalizeHist(src[, dst]) -> dst
@spec equalizeHist(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Equalizes the histogram of a grayscale image.

Positional Arguments
  • src: Evision.Mat.t().

    Source 8-bit single channel image.

Return
  • dst: Evision.Mat.t().

    Destination image of the same size and type as src .

The function equalizes the histogram of the input image using the following algorithm:

  • Calculate the histogram \f$H\f$ for src .

  • Normalize the histogram so that the sum of histogram bins is 255.

  • Compute the integral of the histogram: \f[H'_i = \sum _{0 \le j < i} H(j)\f]

  • Transform the image using \f$H'\f$ as a look-up table: \f$\texttt{dst}(x,y) = H'(\texttt{src}(x,y))\f$

The algorithm normalizes the brightness and increases the contrast of the image.

Python prototype (for reference only):

equalizeHist(src[, dst]) -> dst

Erodes an image by using a specific structuring element.

Positional Arguments
  • src: Evision.Mat.t().

    input image; the number of channels can be arbitrary, but the depth should be one of CV_8U, CV_16U, CV_16S, CV_32F or CV_64F.

  • kernel: Evision.Mat.t().

    structuring element used for erosion; if element=Mat(), a 3 x 3 rectangular structuring element is used. Kernel can be created using #getStructuringElement.

Keyword Arguments
  • anchor: Point.

    position of the anchor within the element; default value (-1, -1) means that the anchor is at the element center.

  • iterations: int.

    number of times erosion is applied.

  • borderType: int.

    pixel extrapolation method, see #BorderTypes. #BORDER_WRAP is not supported.

  • borderValue: Scalar.

    border value in case of a constant border

Return
  • dst: Evision.Mat.t().

    output image of the same size and type as src.

The function erodes the source image using the specified structuring element that determines the shape of a pixel neighborhood over which the minimum is taken: \f[\texttt{dst} (x,y) = \min _{(x',y'): \, \texttt{element} (x',y') \ne0 } \texttt{src} (x+x',y+y')\f] The function supports the in-place mode. Erosion can be applied several ( iterations ) times. In case of multi-channel images, each channel is processed independently. @sa dilate, morphologyEx, getStructuringElement

Python prototype (for reference only):

erode(src, kernel[, dst[, anchor[, iterations[, borderType[, borderValue]]]]]) -> dst
Link to this function

erode(src, kernel, opts)

View Source
@spec erode(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Erodes an image by using a specific structuring element.

Positional Arguments
  • src: Evision.Mat.t().

    input image; the number of channels can be arbitrary, but the depth should be one of CV_8U, CV_16U, CV_16S, CV_32F or CV_64F.

  • kernel: Evision.Mat.t().

    structuring element used for erosion; if element=Mat(), a 3 x 3 rectangular structuring element is used. Kernel can be created using #getStructuringElement.

Keyword Arguments
  • anchor: Point.

    position of the anchor within the element; default value (-1, -1) means that the anchor is at the element center.

  • iterations: int.

    number of times erosion is applied.

  • borderType: int.

    pixel extrapolation method, see #BorderTypes. #BORDER_WRAP is not supported.

  • borderValue: Scalar.

    border value in case of a constant border

Return
  • dst: Evision.Mat.t().

    output image of the same size and type as src.

The function erodes the source image using the specified structuring element that determines the shape of a pixel neighborhood over which the minimum is taken: \f[\texttt{dst} (x,y) = \min _{(x',y'): \, \texttt{element} (x',y') \ne0 } \texttt{src} (x+x',y+y')\f] The function supports the in-place mode. Erosion can be applied several ( iterations ) times. In case of multi-channel images, each channel is processed independently. @sa dilate, morphologyEx, getStructuringElement

Python prototype (for reference only):

erode(src, kernel[, dst[, anchor[, iterations[, borderType[, borderValue]]]]]) -> dst
Link to this function

estimateAffine2D(from, to)

View Source
@spec estimateAffine2D(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in()) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Computes an optimal affine transformation between two 2D point sets.

Positional Arguments
  • from: Evision.Mat.t().

    First input 2D point set containing \f$(X,Y)\f$.

  • to: Evision.Mat.t().

    Second input 2D point set containing \f$(x,y)\f$.

Keyword Arguments
  • method: int.

    Robust method used to compute transformation. The following methods are possible:

    • @ref RANSAC - RANSAC-based robust method
    • @ref LMEDS - Least-Median robust method RANSAC is the default method.
  • ransacReprojThreshold: double.

    Maximum reprojection error in the RANSAC algorithm to consider a point as an inlier. Applies only to RANSAC.

  • maxIters: size_t.

    The maximum number of robust method iterations.

  • confidence: double.

    Confidence level, between 0 and 1, for the estimated transformation. Anything between 0.95 and 0.99 is usually good enough. Values too close to 1 can slow down the estimation significantly. Values lower than 0.8-0.9 can result in an incorrectly estimated transformation.

  • refineIters: size_t.

    Maximum number of iterations of refining algorithm (Levenberg-Marquardt). Passing 0 will disable refining, so the output matrix will be output of robust method.

Return
  • retval: Evision.Mat.t()

  • inliers: Evision.Mat.t().

    Output vector indicating which points are inliers (1-inlier, 0-outlier).

It computes \f[ \begin{bmatrix} x\\ y\\ \end{bmatrix} \begin{bmatrix} a_{11} & a_{12}\\ a_{21} & a_{22}\\ \end{bmatrix} \begin{bmatrix} X\\ Y\\ \end{bmatrix} + \begin{bmatrix} b_1\\ b_2\\ \end{bmatrix} \f]

@return Output 2D affine transformation matrix \f$2 \times 3\f$ or empty matrix if transformation could not be estimated. The returned matrix has the following form: \f[ \begin{bmatrix} a_{11} & a_{12} & b_1\\ a_{21} & a_{22} & b_2\\ \end{bmatrix} \f] The function estimates an optimal 2D affine transformation between two 2D point sets using the selected robust algorithm. The computed transformation is then refined further (using only inliers) with the Levenberg-Marquardt method to reduce the re-projection error even more. Note: The RANSAC method can handle practically any ratio of outliers but needs a threshold to distinguish inliers from outliers. The method LMeDS does not need any threshold but it works correctly only when there are more than 50% of inliers. @sa estimateAffinePartial2D, getAffineTransform

Python prototype (for reference only):

estimateAffine2D(from, to[, inliers[, method[, ransacReprojThreshold[, maxIters[, confidence[, refineIters]]]]]]) -> retval, inliers
Link to this function

estimateAffine2D(from, to, opts)

View Source
@spec estimateAffine2D(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}
@spec estimateAffine2D(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.UsacParams.t()
) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Variant 1:

estimateAffine2D

Positional Arguments
  • pts1: Evision.Mat.t()
  • pts2: Evision.Mat.t()
  • params: Evision.UsacParams.t()
Return
  • retval: Evision.Mat.t()
  • inliers: Evision.Mat.t().

Python prototype (for reference only):

estimateAffine2D(pts1, pts2, params[, inliers]) -> retval, inliers

Variant 2:

Computes an optimal affine transformation between two 2D point sets.

Positional Arguments
  • from: Evision.Mat.t().

    First input 2D point set containing \f$(X,Y)\f$.

  • to: Evision.Mat.t().

    Second input 2D point set containing \f$(x,y)\f$.

Keyword Arguments
  • method: int.

    Robust method used to compute transformation. The following methods are possible:

    • @ref RANSAC - RANSAC-based robust method
    • @ref LMEDS - Least-Median robust method RANSAC is the default method.
  • ransacReprojThreshold: double.

    Maximum reprojection error in the RANSAC algorithm to consider a point as an inlier. Applies only to RANSAC.

  • maxIters: size_t.

    The maximum number of robust method iterations.

  • confidence: double.

    Confidence level, between 0 and 1, for the estimated transformation. Anything between 0.95 and 0.99 is usually good enough. Values too close to 1 can slow down the estimation significantly. Values lower than 0.8-0.9 can result in an incorrectly estimated transformation.

  • refineIters: size_t.

    Maximum number of iterations of refining algorithm (Levenberg-Marquardt). Passing 0 will disable refining, so the output matrix will be output of robust method.

Return
  • retval: Evision.Mat.t()

  • inliers: Evision.Mat.t().

    Output vector indicating which points are inliers (1-inlier, 0-outlier).

It computes \f[ \begin{bmatrix} x\\ y\\ \end{bmatrix} \begin{bmatrix} a_{11} & a_{12}\\ a_{21} & a_{22}\\ \end{bmatrix} \begin{bmatrix} X\\ Y\\ \end{bmatrix} + \begin{bmatrix} b_1\\ b_2\\ \end{bmatrix} \f]

@return Output 2D affine transformation matrix \f$2 \times 3\f$ or empty matrix if transformation could not be estimated. The returned matrix has the following form: \f[ \begin{bmatrix} a_{11} & a_{12} & b_1\\ a_{21} & a_{22} & b_2\\ \end{bmatrix} \f] The function estimates an optimal 2D affine transformation between two 2D point sets using the selected robust algorithm. The computed transformation is then refined further (using only inliers) with the Levenberg-Marquardt method to reduce the re-projection error even more. Note: The RANSAC method can handle practically any ratio of outliers but needs a threshold to distinguish inliers from outliers. The method LMeDS does not need any threshold but it works correctly only when there are more than 50% of inliers. @sa estimateAffinePartial2D, getAffineTransform

Python prototype (for reference only):

estimateAffine2D(from, to[, inliers[, method[, ransacReprojThreshold[, maxIters[, confidence[, refineIters]]]]]]) -> retval, inliers
Link to this function

estimateAffine2D(pts1, pts2, params, opts)

View Source
@spec estimateAffine2D(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.UsacParams.t(),
  [{atom(), term()}, ...] | nil
) :: {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

estimateAffine2D

Positional Arguments
  • pts1: Evision.Mat.t()
  • pts2: Evision.Mat.t()
  • params: Evision.UsacParams.t()
Return
  • retval: Evision.Mat.t()
  • inliers: Evision.Mat.t().

Python prototype (for reference only):

estimateAffine2D(pts1, pts2, params[, inliers]) -> retval, inliers
Link to this function

estimateAffine3D(src, dst)

View Source
@spec estimateAffine3D(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in()) ::
  {Evision.Mat.t(), number()} | {:error, String.t()}

Computes an optimal affine transformation between two 3D point sets.

Positional Arguments
  • src: Evision.Mat.t().

    First input 3D point set.

  • dst: Evision.Mat.t().

    Second input 3D point set.

Keyword Arguments
  • force_rotation: bool.

    If true, the returned rotation will never be a reflection. This might be unwanted, e.g. when optimizing a transform between a right- and a left-handed coordinate system.

Return
  • retval: Evision.Mat.t()

  • scale: double*.

    If null is passed, the scale parameter c will be assumed to be 1.0. Else the pointed-to variable will be set to the optimal scale.

It computes \f$R,s,t\f$ minimizing \f$\sum{i} dst_i - c \cdot R \cdot src_i \f$ where \f$R\f$ is a 3x3 rotation matrix, \f$t\f$ is a 3x1 translation vector and \f$s\f$ is a scalar size value. This is an implementation of the algorithm by Umeyama \cite umeyama1991least . The estimated affine transform has a homogeneous scale which is a subclass of affine transformations with 7 degrees of freedom. The paired point sets need to comprise at least 3 points each. @return 3D affine transformation matrix \f$3 \times 4\f$ of the form \f[T = \begin{bmatrix} R & t\\ \end{bmatrix} \f]

Python prototype (for reference only):

estimateAffine3D(src, dst[, force_rotation]) -> retval, scale
Link to this function

estimateAffine3D(src, dst, opts)

View Source
@spec estimateAffine3D(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: {Evision.Mat.t(), number()} | {:error, String.t()}

Computes an optimal affine transformation between two 3D point sets.

Positional Arguments
  • src: Evision.Mat.t().

    First input 3D point set.

  • dst: Evision.Mat.t().

    Second input 3D point set.

Keyword Arguments
  • force_rotation: bool.

    If true, the returned rotation will never be a reflection. This might be unwanted, e.g. when optimizing a transform between a right- and a left-handed coordinate system.

Return
  • retval: Evision.Mat.t()

  • scale: double*.

    If null is passed, the scale parameter c will be assumed to be 1.0. Else the pointed-to variable will be set to the optimal scale.

It computes \f$R,s,t\f$ minimizing \f$\sum{i} dst_i - c \cdot R \cdot src_i \f$ where \f$R\f$ is a 3x3 rotation matrix, \f$t\f$ is a 3x1 translation vector and \f$s\f$ is a scalar size value. This is an implementation of the algorithm by Umeyama \cite umeyama1991least . The estimated affine transform has a homogeneous scale which is a subclass of affine transformations with 7 degrees of freedom. The paired point sets need to comprise at least 3 points each. @return 3D affine transformation matrix \f$3 \times 4\f$ of the form \f[T = \begin{bmatrix} R & t\\ \end{bmatrix} \f]

Python prototype (for reference only):

estimateAffine3D(src, dst[, force_rotation]) -> retval, scale
Link to this function

estimateAffinePartial2D(from, to)

View Source
@spec estimateAffinePartial2D(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in()) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Computes an optimal limited affine transformation with 4 degrees of freedom between two 2D point sets.

Positional Arguments
  • from: Evision.Mat.t().

    First input 2D point set.

  • to: Evision.Mat.t().

    Second input 2D point set.

Keyword Arguments
  • method: int.

    Robust method used to compute transformation. The following methods are possible:

    • @ref RANSAC - RANSAC-based robust method
    • @ref LMEDS - Least-Median robust method RANSAC is the default method.
  • ransacReprojThreshold: double.

    Maximum reprojection error in the RANSAC algorithm to consider a point as an inlier. Applies only to RANSAC.

  • maxIters: size_t.

    The maximum number of robust method iterations.

  • confidence: double.

    Confidence level, between 0 and 1, for the estimated transformation. Anything between 0.95 and 0.99 is usually good enough. Values too close to 1 can slow down the estimation significantly. Values lower than 0.8-0.9 can result in an incorrectly estimated transformation.

  • refineIters: size_t.

    Maximum number of iterations of refining algorithm (Levenberg-Marquardt). Passing 0 will disable refining, so the output matrix will be output of robust method.

Return
  • retval: Evision.Mat.t()

  • inliers: Evision.Mat.t().

    Output vector indicating which points are inliers.

@return Output 2D affine transformation (4 degrees of freedom) matrix \f$2 \times 3\f$ or empty matrix if transformation could not be estimated. The function estimates an optimal 2D affine transformation with 4 degrees of freedom limited to combinations of translation, rotation, and uniform scaling. Uses the selected algorithm for robust estimation. The computed transformation is then refined further (using only inliers) with the Levenberg-Marquardt method to reduce the re-projection error even more. Estimated transformation matrix is: \f[ \begin{bmatrix} \cos(\theta) \cdot s & -\sin(\theta) \cdot s & t_x \\ \sin(\theta) \cdot s & \cos(\theta) \cdot s & t_y \end{bmatrix} \f] Where \f$ \theta \f$ is the rotation angle, \f$ s \f$ the scaling factor and \f$ t_x, t_y \f$ are translations in \f$ x, y \f$ axes respectively. Note: The RANSAC method can handle practically any ratio of outliers but need a threshold to distinguish inliers from outliers. The method LMeDS does not need any threshold but it works correctly only when there are more than 50% of inliers. @sa estimateAffine2D, getAffineTransform

Python prototype (for reference only):

estimateAffinePartial2D(from, to[, inliers[, method[, ransacReprojThreshold[, maxIters[, confidence[, refineIters]]]]]]) -> retval, inliers
Link to this function

estimateAffinePartial2D(from, to, opts)

View Source
@spec estimateAffinePartial2D(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Computes an optimal limited affine transformation with 4 degrees of freedom between two 2D point sets.

Positional Arguments
  • from: Evision.Mat.t().

    First input 2D point set.

  • to: Evision.Mat.t().

    Second input 2D point set.

Keyword Arguments
  • method: int.

    Robust method used to compute transformation. The following methods are possible:

    • @ref RANSAC - RANSAC-based robust method
    • @ref LMEDS - Least-Median robust method RANSAC is the default method.
  • ransacReprojThreshold: double.

    Maximum reprojection error in the RANSAC algorithm to consider a point as an inlier. Applies only to RANSAC.

  • maxIters: size_t.

    The maximum number of robust method iterations.

  • confidence: double.

    Confidence level, between 0 and 1, for the estimated transformation. Anything between 0.95 and 0.99 is usually good enough. Values too close to 1 can slow down the estimation significantly. Values lower than 0.8-0.9 can result in an incorrectly estimated transformation.

  • refineIters: size_t.

    Maximum number of iterations of refining algorithm (Levenberg-Marquardt). Passing 0 will disable refining, so the output matrix will be output of robust method.

Return
  • retval: Evision.Mat.t()

  • inliers: Evision.Mat.t().

    Output vector indicating which points are inliers.

@return Output 2D affine transformation (4 degrees of freedom) matrix \f$2 \times 3\f$ or empty matrix if transformation could not be estimated. The function estimates an optimal 2D affine transformation with 4 degrees of freedom limited to combinations of translation, rotation, and uniform scaling. Uses the selected algorithm for robust estimation. The computed transformation is then refined further (using only inliers) with the Levenberg-Marquardt method to reduce the re-projection error even more. Estimated transformation matrix is: \f[ \begin{bmatrix} \cos(\theta) \cdot s & -\sin(\theta) \cdot s & t_x \\ \sin(\theta) \cdot s & \cos(\theta) \cdot s & t_y \end{bmatrix} \f] Where \f$ \theta \f$ is the rotation angle, \f$ s \f$ the scaling factor and \f$ t_x, t_y \f$ are translations in \f$ x, y \f$ axes respectively. Note: The RANSAC method can handle practically any ratio of outliers but need a threshold to distinguish inliers from outliers. The method LMeDS does not need any threshold but it works correctly only when there are more than 50% of inliers. @sa estimateAffine2D, getAffineTransform

Python prototype (for reference only):

estimateAffinePartial2D(from, to[, inliers[, method[, ransacReprojThreshold[, maxIters[, confidence[, refineIters]]]]]]) -> retval, inliers
Link to this function

estimateChessboardSharpness(image, patternSize, corners)

View Source
@spec estimateChessboardSharpness(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  Evision.Mat.maybe_mat_in()
) ::
  {{number()}
   | {number(), number()}
   | {number(), number(), number()}
   | {number(), number(), number(), number()}, Evision.Mat.t()}
  | {:error, String.t()}

Estimates the sharpness of a detected chessboard.

Positional Arguments
  • image: Evision.Mat.t().

    Gray image used to find chessboard corners

  • patternSize: Size.

    Size of a found chessboard pattern

  • corners: Evision.Mat.t().

    Corners found by #findChessboardCornersSB

Keyword Arguments
  • rise_distance: float.

    Rise distance 0.8 means 10% ... 90% of the final signal strength

  • vertical: bool.

    By default edge responses for horizontal lines are calculated

Return
  • retval: Scalar

  • sharpness: Evision.Mat.t().

    Optional output array with a sharpness value for calculated edge responses (see description)

Image sharpness, as well as brightness, are a critical parameter for accuracte camera calibration. For accessing these parameters for filtering out problematic calibraiton images, this method calculates edge profiles by traveling from black to white chessboard cell centers. Based on this, the number of pixels is calculated required to transit from black to white. This width of the transition area is a good indication of how sharp the chessboard is imaged and should be below ~3.0 pixels.

The optional sharpness array is of type CV_32FC1 and has for each calculated profile one row with the following five entries: 0 = x coordinate of the underlying edge in the image 1 = y coordinate of the underlying edge in the image 2 = width of the transition area (sharpness) 3 = signal strength in the black cell (min brightness) 4 = signal strength in the white cell (max brightness) @return Scalar(average sharpness, average min brightness, average max brightness,0)

Python prototype (for reference only):

estimateChessboardSharpness(image, patternSize, corners[, rise_distance[, vertical[, sharpness]]]) -> retval, sharpness
Link to this function

estimateChessboardSharpness(image, patternSize, corners, opts)

View Source
@spec estimateChessboardSharpness(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  {{number()}
   | {number(), number()}
   | {number(), number(), number()}
   | {number(), number(), number(), number()}, Evision.Mat.t()}
  | {:error, String.t()}

Estimates the sharpness of a detected chessboard.

Positional Arguments
  • image: Evision.Mat.t().

    Gray image used to find chessboard corners

  • patternSize: Size.

    Size of a found chessboard pattern

  • corners: Evision.Mat.t().

    Corners found by #findChessboardCornersSB

Keyword Arguments
  • rise_distance: float.

    Rise distance 0.8 means 10% ... 90% of the final signal strength

  • vertical: bool.

    By default edge responses for horizontal lines are calculated

Return
  • retval: Scalar

  • sharpness: Evision.Mat.t().

    Optional output array with a sharpness value for calculated edge responses (see description)

Image sharpness, as well as brightness, are a critical parameter for accuracte camera calibration. For accessing these parameters for filtering out problematic calibraiton images, this method calculates edge profiles by traveling from black to white chessboard cell centers. Based on this, the number of pixels is calculated required to transit from black to white. This width of the transition area is a good indication of how sharp the chessboard is imaged and should be below ~3.0 pixels.

The optional sharpness array is of type CV_32FC1 and has for each calculated profile one row with the following five entries: 0 = x coordinate of the underlying edge in the image 1 = y coordinate of the underlying edge in the image 2 = width of the transition area (sharpness) 3 = signal strength in the black cell (min brightness) 4 = signal strength in the white cell (max brightness) @return Scalar(average sharpness, average min brightness, average max brightness,0)

Python prototype (for reference only):

estimateChessboardSharpness(image, patternSize, corners[, rise_distance[, vertical[, sharpness]]]) -> retval, sharpness
Link to this function

estimateTranslation3D(src, dst)

View Source
@spec estimateTranslation3D(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in()) ::
  {integer(), Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Computes an optimal translation between two 3D point sets.

Positional Arguments
  • src: Evision.Mat.t().

    First input 3D point set containing \f$(X,Y,Z)\f$.

  • dst: Evision.Mat.t().

    Second input 3D point set containing \f$(x,y,z)\f$.

Keyword Arguments
  • ransacThreshold: double.

    Maximum reprojection error in the RANSAC algorithm to consider a point as an inlier.

  • confidence: double.

    Confidence level, between 0 and 1, for the estimated transformation. Anything between 0.95 and 0.99 is usually good enough. Values too close to 1 can slow down the estimation significantly. Values lower than 0.8-0.9 can result in an incorrectly estimated transformation.

Return
  • retval: int

  • out: Evision.Mat.t().

    Output 3D translation vector \f$3 \times 1\f$ of the form \f[ \begin{bmatrix} b_1 \\ b_2 \\ b_3 \\ \end{bmatrix} \f]

  • inliers: Evision.Mat.t().

    Output vector indicating which points are inliers (1-inlier, 0-outlier).

It computes \f[ \begin{bmatrix} x\\ y\\ z\\ \end{bmatrix} \begin{bmatrix} X\\ Y\\ Z\\ \end{bmatrix} + \begin{bmatrix} b_1\\ b_2\\ b_3\\ \end{bmatrix} \f]

The function estimates an optimal 3D translation between two 3D point sets using the RANSAC algorithm.

Python prototype (for reference only):

estimateTranslation3D(src, dst[, out[, inliers[, ransacThreshold[, confidence]]]]) -> retval, out, inliers
Link to this function

estimateTranslation3D(src, dst, opts)

View Source
@spec estimateTranslation3D(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: {integer(), Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Computes an optimal translation between two 3D point sets.

Positional Arguments
  • src: Evision.Mat.t().

    First input 3D point set containing \f$(X,Y,Z)\f$.

  • dst: Evision.Mat.t().

    Second input 3D point set containing \f$(x,y,z)\f$.

Keyword Arguments
  • ransacThreshold: double.

    Maximum reprojection error in the RANSAC algorithm to consider a point as an inlier.

  • confidence: double.

    Confidence level, between 0 and 1, for the estimated transformation. Anything between 0.95 and 0.99 is usually good enough. Values too close to 1 can slow down the estimation significantly. Values lower than 0.8-0.9 can result in an incorrectly estimated transformation.

Return
  • retval: int

  • out: Evision.Mat.t().

    Output 3D translation vector \f$3 \times 1\f$ of the form \f[ \begin{bmatrix} b_1 \\ b_2 \\ b_3 \\ \end{bmatrix} \f]

  • inliers: Evision.Mat.t().

    Output vector indicating which points are inliers (1-inlier, 0-outlier).

It computes \f[ \begin{bmatrix} x\\ y\\ z\\ \end{bmatrix} \begin{bmatrix} X\\ Y\\ Z\\ \end{bmatrix} + \begin{bmatrix} b_1\\ b_2\\ b_3\\ \end{bmatrix} \f]

The function estimates an optimal 3D translation between two 3D point sets using the RANSAC algorithm.

Python prototype (for reference only):

estimateTranslation3D(src, dst[, out[, inliers[, ransacThreshold[, confidence]]]]) -> retval, out, inliers
@spec exp(Evision.Mat.maybe_mat_in()) :: Evision.Mat.t() | {:error, String.t()}

Calculates the exponent of every array element.

Positional Arguments
  • src: Evision.Mat.t().

    input array.

Return
  • dst: Evision.Mat.t().

    output array of the same size and type as src.

The function cv::exp calculates the exponent of every element of the input array: \f[\texttt{dst} [I] = e^{ src(I) }\f] The maximum relative error is about 7e-6 for single-precision input and less than 1e-10 for double-precision input. Currently, the function converts denormalized values to zeros on output. Special values (NaN, Inf) are not handled. @sa log , cartToPolar , polarToCart , phase , pow , sqrt , magnitude

Python prototype (for reference only):

exp(src[, dst]) -> dst
@spec exp(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Calculates the exponent of every array element.

Positional Arguments
  • src: Evision.Mat.t().

    input array.

Return
  • dst: Evision.Mat.t().

    output array of the same size and type as src.

The function cv::exp calculates the exponent of every element of the input array: \f[\texttt{dst} [I] = e^{ src(I) }\f] The maximum relative error is about 7e-6 for single-precision input and less than 1e-10 for double-precision input. Currently, the function converts denormalized values to zeros on output. Special values (NaN, Inf) are not handled. @sa log , cartToPolar , polarToCart , phase , pow , sqrt , magnitude

Python prototype (for reference only):

exp(src[, dst]) -> dst
Link to this function

extractChannel(src, coi)

View Source
@spec extractChannel(Evision.Mat.maybe_mat_in(), integer()) ::
  Evision.Mat.t() | {:error, String.t()}

Extracts a single channel from src (coi is 0-based index)

Positional Arguments
  • src: Evision.Mat.t().

    input array

  • coi: int.

    index of channel to extract

Return
  • dst: Evision.Mat.t().

    output array

@sa mixChannels, split

Python prototype (for reference only):

extractChannel(src, coi[, dst]) -> dst
Link to this function

extractChannel(src, coi, opts)

View Source
@spec extractChannel(
  Evision.Mat.maybe_mat_in(),
  integer(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Extracts a single channel from src (coi is 0-based index)

Positional Arguments
  • src: Evision.Mat.t().

    input array

  • coi: int.

    index of channel to extract

Return
  • dst: Evision.Mat.t().

    output array

@sa mixChannels, split

Python prototype (for reference only):

extractChannel(src, coi[, dst]) -> dst
@spec fastAtan2(number(), number()) :: number() | {:error, String.t()}

Calculates the angle of a 2D vector in degrees.

Positional Arguments
  • y: float.

    y-coordinate of the vector.

  • x: float.

    x-coordinate of the vector.

Return
  • retval: float

The function fastAtan2 calculates the full-range angle of an input 2D vector. The angle is measured in degrees and varies from 0 to 360 degrees. The accuracy is about 0.3 degrees.

Python prototype (for reference only):

fastAtan2(y, x) -> retval
Link to this function

fastNlMeansDenoising(src)

View Source
@spec fastNlMeansDenoising(Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

Perform image denoising using Non-local Means Denoising algorithm http://www.ipol.im/pub/algo/bcm_non_local_means_denoising/ with several computational optimizations. Noise expected to be a gaussian white noise

Positional Arguments
  • src: Evision.Mat.t().

    Input 8-bit 1-channel, 2-channel, 3-channel or 4-channel image.

Keyword Arguments
  • h: float.

    Parameter regulating filter strength. Big h value perfectly removes noise but also removes image details, smaller h value preserves details but also preserves some noise

  • templateWindowSize: int.

    Size in pixels of the template patch that is used to compute weights. Should be odd. Recommended value 7 pixels

  • searchWindowSize: int.

    Size in pixels of the window that is used to compute weighted average for given pixel. Should be odd. Affect performance linearly: greater searchWindowsSize - greater denoising time. Recommended value 21 pixels

Return
  • dst: Evision.Mat.t().

    Output image with the same size and type as src .

This function expected to be applied to grayscale images. For colored images look at fastNlMeansDenoisingColored. Advanced usage of this functions can be manual denoising of colored image in different colorspaces. Such approach is used in fastNlMeansDenoisingColored by converting image to CIELAB colorspace and then separately denoise L and AB components with different h parameter.

Python prototype (for reference only):

fastNlMeansDenoising(src[, dst[, h[, templateWindowSize[, searchWindowSize]]]]) -> dst
Link to this function

fastNlMeansDenoising(src, opts)

View Source
@spec fastNlMeansDenoising(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}
@spec fastNlMeansDenoising(Evision.Mat.maybe_mat_in(), [number()]) ::
  Evision.Mat.t() | {:error, String.t()}

Variant 1:

Perform image denoising using Non-local Means Denoising algorithm http://www.ipol.im/pub/algo/bcm_non_local_means_denoising/ with several computational optimizations. Noise expected to be a gaussian white noise

Positional Arguments
  • src: Evision.Mat.t().

    Input 8-bit or 16-bit (only with NORM_L1) 1-channel, 2-channel, 3-channel or 4-channel image.

  • h: [float].

    Array of parameters regulating filter strength, either one parameter applied to all channels or one per channel in dst. Big h value perfectly removes noise but also removes image details, smaller h value preserves details but also preserves some noise

Keyword Arguments
  • templateWindowSize: int.

    Size in pixels of the template patch that is used to compute weights. Should be odd. Recommended value 7 pixels

  • searchWindowSize: int.

    Size in pixels of the window that is used to compute weighted average for given pixel. Should be odd. Affect performance linearly: greater searchWindowsSize - greater denoising time. Recommended value 21 pixels

  • normType: int.

    Type of norm used for weight calculation. Can be either NORM_L2 or NORM_L1

Return
  • dst: Evision.Mat.t().

    Output image with the same size and type as src .

This function expected to be applied to grayscale images. For colored images look at fastNlMeansDenoisingColored. Advanced usage of this functions can be manual denoising of colored image in different colorspaces. Such approach is used in fastNlMeansDenoisingColored by converting image to CIELAB colorspace and then separately denoise L and AB components with different h parameter.

Python prototype (for reference only):

fastNlMeansDenoising(src, h[, dst[, templateWindowSize[, searchWindowSize[, normType]]]]) -> dst

Variant 2:

Perform image denoising using Non-local Means Denoising algorithm http://www.ipol.im/pub/algo/bcm_non_local_means_denoising/ with several computational optimizations. Noise expected to be a gaussian white noise

Positional Arguments
  • src: Evision.Mat.t().

    Input 8-bit 1-channel, 2-channel, 3-channel or 4-channel image.

Keyword Arguments
  • h: float.

    Parameter regulating filter strength. Big h value perfectly removes noise but also removes image details, smaller h value preserves details but also preserves some noise

  • templateWindowSize: int.

    Size in pixels of the template patch that is used to compute weights. Should be odd. Recommended value 7 pixels

  • searchWindowSize: int.

    Size in pixels of the window that is used to compute weighted average for given pixel. Should be odd. Affect performance linearly: greater searchWindowsSize - greater denoising time. Recommended value 21 pixels

Return
  • dst: Evision.Mat.t().

    Output image with the same size and type as src .

This function expected to be applied to grayscale images. For colored images look at fastNlMeansDenoisingColored. Advanced usage of this functions can be manual denoising of colored image in different colorspaces. Such approach is used in fastNlMeansDenoisingColored by converting image to CIELAB colorspace and then separately denoise L and AB components with different h parameter.

Python prototype (for reference only):

fastNlMeansDenoising(src[, dst[, h[, templateWindowSize[, searchWindowSize]]]]) -> dst
Link to this function

fastNlMeansDenoising(src, h, opts)

View Source
@spec fastNlMeansDenoising(
  Evision.Mat.maybe_mat_in(),
  [number()],
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Perform image denoising using Non-local Means Denoising algorithm http://www.ipol.im/pub/algo/bcm_non_local_means_denoising/ with several computational optimizations. Noise expected to be a gaussian white noise

Positional Arguments
  • src: Evision.Mat.t().

    Input 8-bit or 16-bit (only with NORM_L1) 1-channel, 2-channel, 3-channel or 4-channel image.

  • h: [float].

    Array of parameters regulating filter strength, either one parameter applied to all channels or one per channel in dst. Big h value perfectly removes noise but also removes image details, smaller h value preserves details but also preserves some noise

Keyword Arguments
  • templateWindowSize: int.

    Size in pixels of the template patch that is used to compute weights. Should be odd. Recommended value 7 pixels

  • searchWindowSize: int.

    Size in pixels of the window that is used to compute weighted average for given pixel. Should be odd. Affect performance linearly: greater searchWindowsSize - greater denoising time. Recommended value 21 pixels

  • normType: int.

    Type of norm used for weight calculation. Can be either NORM_L2 or NORM_L1

Return
  • dst: Evision.Mat.t().

    Output image with the same size and type as src .

This function expected to be applied to grayscale images. For colored images look at fastNlMeansDenoisingColored. Advanced usage of this functions can be manual denoising of colored image in different colorspaces. Such approach is used in fastNlMeansDenoisingColored by converting image to CIELAB colorspace and then separately denoise L and AB components with different h parameter.

Python prototype (for reference only):

fastNlMeansDenoising(src, h[, dst[, templateWindowSize[, searchWindowSize[, normType]]]]) -> dst
Link to this function

fastNlMeansDenoisingColored(src)

View Source
@spec fastNlMeansDenoisingColored(Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

Modification of fastNlMeansDenoising function for colored images

Positional Arguments
  • src: Evision.Mat.t().

    Input 8-bit 3-channel image.

Keyword Arguments
  • h: float.

    Parameter regulating filter strength for luminance component. Bigger h value perfectly removes noise but also removes image details, smaller h value preserves details but also preserves some noise

  • hColor: float.

    The same as h but for color components. For most images value equals 10 will be enough to remove colored noise and do not distort colors

  • templateWindowSize: int.

    Size in pixels of the template patch that is used to compute weights. Should be odd. Recommended value 7 pixels

  • searchWindowSize: int.

    Size in pixels of the window that is used to compute weighted average for given pixel. Should be odd. Affect performance linearly: greater searchWindowsSize - greater denoising time. Recommended value 21 pixels

Return
  • dst: Evision.Mat.t().

    Output image with the same size and type as src .

The function converts image to CIELAB colorspace and then separately denoise L and AB components with given h parameters using fastNlMeansDenoising function.

Python prototype (for reference only):

fastNlMeansDenoisingColored(src[, dst[, h[, hColor[, templateWindowSize[, searchWindowSize]]]]]) -> dst
Link to this function

fastNlMeansDenoisingColored(src, opts)

View Source
@spec fastNlMeansDenoisingColored(
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Modification of fastNlMeansDenoising function for colored images

Positional Arguments
  • src: Evision.Mat.t().

    Input 8-bit 3-channel image.

Keyword Arguments
  • h: float.

    Parameter regulating filter strength for luminance component. Bigger h value perfectly removes noise but also removes image details, smaller h value preserves details but also preserves some noise

  • hColor: float.

    The same as h but for color components. For most images value equals 10 will be enough to remove colored noise and do not distort colors

  • templateWindowSize: int.

    Size in pixels of the template patch that is used to compute weights. Should be odd. Recommended value 7 pixels

  • searchWindowSize: int.

    Size in pixels of the window that is used to compute weighted average for given pixel. Should be odd. Affect performance linearly: greater searchWindowsSize - greater denoising time. Recommended value 21 pixels

Return
  • dst: Evision.Mat.t().

    Output image with the same size and type as src .

The function converts image to CIELAB colorspace and then separately denoise L and AB components with given h parameters using fastNlMeansDenoising function.

Python prototype (for reference only):

fastNlMeansDenoisingColored(src[, dst[, h[, hColor[, templateWindowSize[, searchWindowSize]]]]]) -> dst
Link to this function

fastNlMeansDenoisingColoredMulti(srcImgs, imgToDenoiseIndex, temporalWindowSize)

View Source
@spec fastNlMeansDenoisingColoredMulti(
  [Evision.Mat.maybe_mat_in()],
  integer(),
  integer()
) ::
  Evision.Mat.t() | {:error, String.t()}

Modification of fastNlMeansDenoisingMulti function for colored images sequences

Positional Arguments
  • srcImgs: [Evision.Mat].

    Input 8-bit 3-channel images sequence. All images should have the same type and size.

  • imgToDenoiseIndex: int.

    Target image to denoise index in srcImgs sequence

  • temporalWindowSize: int.

    Number of surrounding images to use for target image denoising. Should be odd. Images from imgToDenoiseIndex - temporalWindowSize / 2 to imgToDenoiseIndex - temporalWindowSize / 2 from srcImgs will be used to denoise srcImgs[imgToDenoiseIndex] image.

Keyword Arguments
  • h: float.

    Parameter regulating filter strength for luminance component. Bigger h value perfectly removes noise but also removes image details, smaller h value preserves details but also preserves some noise.

  • hColor: float.

    The same as h but for color components.

  • templateWindowSize: int.

    Size in pixels of the template patch that is used to compute weights. Should be odd. Recommended value 7 pixels

  • searchWindowSize: int.

    Size in pixels of the window that is used to compute weighted average for given pixel. Should be odd. Affect performance linearly: greater searchWindowsSize - greater denoising time. Recommended value 21 pixels

Return
  • dst: Evision.Mat.t().

    Output image with the same size and type as srcImgs images.

The function converts images to CIELAB colorspace and then separately denoise L and AB components with given h parameters using fastNlMeansDenoisingMulti function.

Python prototype (for reference only):

fastNlMeansDenoisingColoredMulti(srcImgs, imgToDenoiseIndex, temporalWindowSize[, dst[, h[, hColor[, templateWindowSize[, searchWindowSize]]]]]) -> dst
Link to this function

fastNlMeansDenoisingColoredMulti(srcImgs, imgToDenoiseIndex, temporalWindowSize, opts)

View Source
@spec fastNlMeansDenoisingColoredMulti(
  [Evision.Mat.maybe_mat_in()],
  integer(),
  integer(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Modification of fastNlMeansDenoisingMulti function for colored images sequences

Positional Arguments
  • srcImgs: [Evision.Mat].

    Input 8-bit 3-channel images sequence. All images should have the same type and size.

  • imgToDenoiseIndex: int.

    Target image to denoise index in srcImgs sequence

  • temporalWindowSize: int.

    Number of surrounding images to use for target image denoising. Should be odd. Images from imgToDenoiseIndex - temporalWindowSize / 2 to imgToDenoiseIndex - temporalWindowSize / 2 from srcImgs will be used to denoise srcImgs[imgToDenoiseIndex] image.

Keyword Arguments
  • h: float.

    Parameter regulating filter strength for luminance component. Bigger h value perfectly removes noise but also removes image details, smaller h value preserves details but also preserves some noise.

  • hColor: float.

    The same as h but for color components.

  • templateWindowSize: int.

    Size in pixels of the template patch that is used to compute weights. Should be odd. Recommended value 7 pixels

  • searchWindowSize: int.

    Size in pixels of the window that is used to compute weighted average for given pixel. Should be odd. Affect performance linearly: greater searchWindowsSize - greater denoising time. Recommended value 21 pixels

Return
  • dst: Evision.Mat.t().

    Output image with the same size and type as srcImgs images.

The function converts images to CIELAB colorspace and then separately denoise L and AB components with given h parameters using fastNlMeansDenoisingMulti function.

Python prototype (for reference only):

fastNlMeansDenoisingColoredMulti(srcImgs, imgToDenoiseIndex, temporalWindowSize[, dst[, h[, hColor[, templateWindowSize[, searchWindowSize]]]]]) -> dst
Link to this function

fastNlMeansDenoisingMulti(srcImgs, imgToDenoiseIndex, temporalWindowSize)

View Source
@spec fastNlMeansDenoisingMulti([Evision.Mat.maybe_mat_in()], integer(), integer()) ::
  Evision.Mat.t() | {:error, String.t()}

Modification of fastNlMeansDenoising function for images sequence where consecutive images have been captured in small period of time. For example video. This version of the function is for grayscale images or for manual manipulation with colorspaces. See @cite Buades2005DenoisingIS for more details (open access here).

Positional Arguments
  • srcImgs: [Evision.Mat].

    Input 8-bit 1-channel, 2-channel, 3-channel or 4-channel images sequence. All images should have the same type and size.

  • imgToDenoiseIndex: int.

    Target image to denoise index in srcImgs sequence

  • temporalWindowSize: int.

    Number of surrounding images to use for target image denoising. Should be odd. Images from imgToDenoiseIndex - temporalWindowSize / 2 to imgToDenoiseIndex - temporalWindowSize / 2 from srcImgs will be used to denoise srcImgs[imgToDenoiseIndex] image.

Keyword Arguments
  • h: float.

    Parameter regulating filter strength. Bigger h value perfectly removes noise but also removes image details, smaller h value preserves details but also preserves some noise

  • templateWindowSize: int.

    Size in pixels of the template patch that is used to compute weights. Should be odd. Recommended value 7 pixels

  • searchWindowSize: int.

    Size in pixels of the window that is used to compute weighted average for given pixel. Should be odd. Affect performance linearly: greater searchWindowsSize - greater denoising time. Recommended value 21 pixels

Return
  • dst: Evision.Mat.t().

    Output image with the same size and type as srcImgs images.

Python prototype (for reference only):

fastNlMeansDenoisingMulti(srcImgs, imgToDenoiseIndex, temporalWindowSize[, dst[, h[, templateWindowSize[, searchWindowSize]]]]) -> dst
Link to this function

fastNlMeansDenoisingMulti(srcImgs, imgToDenoiseIndex, temporalWindowSize, opts)

View Source
@spec fastNlMeansDenoisingMulti(
  [Evision.Mat.maybe_mat_in()],
  integer(),
  integer(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}
@spec fastNlMeansDenoisingMulti([Evision.Mat.maybe_mat_in()], integer(), integer(), [
  number()
]) ::
  Evision.Mat.t() | {:error, String.t()}

Variant 1:

Modification of fastNlMeansDenoising function for images sequence where consecutive images have been captured in small period of time. For example video. This version of the function is for grayscale images or for manual manipulation with colorspaces. See @cite Buades2005DenoisingIS for more details (open access here).

Positional Arguments
  • srcImgs: [Evision.Mat].

    Input 8-bit or 16-bit (only with NORM_L1) 1-channel, 2-channel, 3-channel or 4-channel images sequence. All images should have the same type and size.

  • imgToDenoiseIndex: int.

    Target image to denoise index in srcImgs sequence

  • temporalWindowSize: int.

    Number of surrounding images to use for target image denoising. Should be odd. Images from imgToDenoiseIndex - temporalWindowSize / 2 to imgToDenoiseIndex - temporalWindowSize / 2 from srcImgs will be used to denoise srcImgs[imgToDenoiseIndex] image.

  • h: [float].

    Array of parameters regulating filter strength, either one parameter applied to all channels or one per channel in dst. Big h value perfectly removes noise but also removes image details, smaller h value preserves details but also preserves some noise

Keyword Arguments
  • templateWindowSize: int.

    Size in pixels of the template patch that is used to compute weights. Should be odd. Recommended value 7 pixels

  • searchWindowSize: int.

    Size in pixels of the window that is used to compute weighted average for given pixel. Should be odd. Affect performance linearly: greater searchWindowsSize - greater denoising time. Recommended value 21 pixels

  • normType: int.

    Type of norm used for weight calculation. Can be either NORM_L2 or NORM_L1

Return
  • dst: Evision.Mat.t().

    Output image with the same size and type as srcImgs images.

Python prototype (for reference only):

fastNlMeansDenoisingMulti(srcImgs, imgToDenoiseIndex, temporalWindowSize, h[, dst[, templateWindowSize[, searchWindowSize[, normType]]]]) -> dst

Variant 2:

Modification of fastNlMeansDenoising function for images sequence where consecutive images have been captured in small period of time. For example video. This version of the function is for grayscale images or for manual manipulation with colorspaces. See @cite Buades2005DenoisingIS for more details (open access here).

Positional Arguments
  • srcImgs: [Evision.Mat].

    Input 8-bit 1-channel, 2-channel, 3-channel or 4-channel images sequence. All images should have the same type and size.

  • imgToDenoiseIndex: int.

    Target image to denoise index in srcImgs sequence

  • temporalWindowSize: int.

    Number of surrounding images to use for target image denoising. Should be odd. Images from imgToDenoiseIndex - temporalWindowSize / 2 to imgToDenoiseIndex - temporalWindowSize / 2 from srcImgs will be used to denoise srcImgs[imgToDenoiseIndex] image.

Keyword Arguments
  • h: float.

    Parameter regulating filter strength. Bigger h value perfectly removes noise but also removes image details, smaller h value preserves details but also preserves some noise

  • templateWindowSize: int.

    Size in pixels of the template patch that is used to compute weights. Should be odd. Recommended value 7 pixels

  • searchWindowSize: int.

    Size in pixels of the window that is used to compute weighted average for given pixel. Should be odd. Affect performance linearly: greater searchWindowsSize - greater denoising time. Recommended value 21 pixels

Return
  • dst: Evision.Mat.t().

    Output image with the same size and type as srcImgs images.

Python prototype (for reference only):

fastNlMeansDenoisingMulti(srcImgs, imgToDenoiseIndex, temporalWindowSize[, dst[, h[, templateWindowSize[, searchWindowSize]]]]) -> dst
Link to this function

fastNlMeansDenoisingMulti(srcImgs, imgToDenoiseIndex, temporalWindowSize, h, opts)

View Source
@spec fastNlMeansDenoisingMulti(
  [Evision.Mat.maybe_mat_in()],
  integer(),
  integer(),
  [number()],
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Modification of fastNlMeansDenoising function for images sequence where consecutive images have been captured in small period of time. For example video. This version of the function is for grayscale images or for manual manipulation with colorspaces. See @cite Buades2005DenoisingIS for more details (open access here).

Positional Arguments
  • srcImgs: [Evision.Mat].

    Input 8-bit or 16-bit (only with NORM_L1) 1-channel, 2-channel, 3-channel or 4-channel images sequence. All images should have the same type and size.

  • imgToDenoiseIndex: int.

    Target image to denoise index in srcImgs sequence

  • temporalWindowSize: int.

    Number of surrounding images to use for target image denoising. Should be odd. Images from imgToDenoiseIndex - temporalWindowSize / 2 to imgToDenoiseIndex - temporalWindowSize / 2 from srcImgs will be used to denoise srcImgs[imgToDenoiseIndex] image.

  • h: [float].

    Array of parameters regulating filter strength, either one parameter applied to all channels or one per channel in dst. Big h value perfectly removes noise but also removes image details, smaller h value preserves details but also preserves some noise

Keyword Arguments
  • templateWindowSize: int.

    Size in pixels of the template patch that is used to compute weights. Should be odd. Recommended value 7 pixels

  • searchWindowSize: int.

    Size in pixels of the window that is used to compute weighted average for given pixel. Should be odd. Affect performance linearly: greater searchWindowsSize - greater denoising time. Recommended value 21 pixels

  • normType: int.

    Type of norm used for weight calculation. Can be either NORM_L2 or NORM_L1

Return
  • dst: Evision.Mat.t().

    Output image with the same size and type as srcImgs images.

Python prototype (for reference only):

fastNlMeansDenoisingMulti(srcImgs, imgToDenoiseIndex, temporalWindowSize, h[, dst[, templateWindowSize[, searchWindowSize[, normType]]]]) -> dst
Link to this function

fillConvexPoly(img, points, color)

View Source
@spec fillConvexPoly(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()}
) :: Evision.Mat.t() | {:error, String.t()}

Fills a convex polygon.

Positional Arguments
  • points: Evision.Mat.t().

    Polygon vertices.

  • color: Scalar.

    Polygon color.

Keyword Arguments
  • lineType: int.

    Type of the polygon boundaries. See #LineTypes

  • shift: int.

    Number of fractional bits in the vertex coordinates.

Return
  • img: Evision.Mat.t().

    Image.

The function cv::fillConvexPoly draws a filled convex polygon. This function is much faster than the function #fillPoly . It can fill not only convex polygons but any monotonic polygon without self-intersections, that is, a polygon whose contour intersects every horizontal line (scan line) twice at the most (though, its top-most and/or the bottom edge could be horizontal).

Python prototype (for reference only):

fillConvexPoly(img, points, color[, lineType[, shift]]) -> img
Link to this function

fillConvexPoly(img, points, color, opts)

View Source
@spec fillConvexPoly(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()},
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Fills a convex polygon.

Positional Arguments
  • points: Evision.Mat.t().

    Polygon vertices.

  • color: Scalar.

    Polygon color.

Keyword Arguments
  • lineType: int.

    Type of the polygon boundaries. See #LineTypes

  • shift: int.

    Number of fractional bits in the vertex coordinates.

Return
  • img: Evision.Mat.t().

    Image.

The function cv::fillConvexPoly draws a filled convex polygon. This function is much faster than the function #fillPoly . It can fill not only convex polygons but any monotonic polygon without self-intersections, that is, a polygon whose contour intersects every horizontal line (scan line) twice at the most (though, its top-most and/or the bottom edge could be horizontal).

Python prototype (for reference only):

fillConvexPoly(img, points, color[, lineType[, shift]]) -> img
Link to this function

fillPoly(img, pts, color)

View Source
@spec fillPoly(
  Evision.Mat.maybe_mat_in(),
  [Evision.Mat.maybe_mat_in()],
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()}
) :: Evision.Mat.t() | {:error, String.t()}

Fills the area bounded by one or more polygons.

Positional Arguments
  • pts: [Evision.Mat].

    Array of polygons where each polygon is represented as an array of points.

  • color: Scalar.

    Polygon color.

Keyword Arguments
  • lineType: int.

    Type of the polygon boundaries. See #LineTypes

  • shift: int.

    Number of fractional bits in the vertex coordinates.

  • offset: Point.

    Optional offset of all points of the contours.

Return
  • img: Evision.Mat.t().

    Image.

The function cv::fillPoly fills an area bounded by several polygonal contours. The function can fill complex areas, for example, areas with holes, contours with self-intersections (some of their parts), and so forth.

Python prototype (for reference only):

fillPoly(img, pts, color[, lineType[, shift[, offset]]]) -> img
Link to this function

fillPoly(img, pts, color, opts)

View Source
@spec fillPoly(
  Evision.Mat.maybe_mat_in(),
  [Evision.Mat.maybe_mat_in()],
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()},
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Fills the area bounded by one or more polygons.

Positional Arguments
  • pts: [Evision.Mat].

    Array of polygons where each polygon is represented as an array of points.

  • color: Scalar.

    Polygon color.

Keyword Arguments
  • lineType: int.

    Type of the polygon boundaries. See #LineTypes

  • shift: int.

    Number of fractional bits in the vertex coordinates.

  • offset: Point.

    Optional offset of all points of the contours.

Return
  • img: Evision.Mat.t().

    Image.

The function cv::fillPoly fills an area bounded by several polygonal contours. The function can fill complex areas, for example, areas with holes, contours with self-intersections (some of their parts), and so forth.

Python prototype (for reference only):

fillPoly(img, pts, color[, lineType[, shift[, offset]]]) -> img
Link to this function

filter2D(src, ddepth, kernel)

View Source

Convolves an image with the kernel.

Positional Arguments
  • src: Evision.Mat.t().

    input image.

  • ddepth: int.

    desired depth of the destination image, see @ref filter_depths "combinations"

  • kernel: Evision.Mat.t().

    convolution kernel (or rather a correlation kernel), a single-channel floating point matrix; if you want to apply different kernels to different channels, split the image into separate color planes using split and process them individually.

Keyword Arguments
  • anchor: Point.

    anchor of the kernel that indicates the relative position of a filtered point within the kernel; the anchor should lie within the kernel; default value (-1,-1) means that the anchor is at the kernel center.

  • delta: double.

    optional value added to the filtered pixels before storing them in dst.

  • borderType: int.

    pixel extrapolation method, see #BorderTypes. #BORDER_WRAP is not supported.

Return
  • dst: Evision.Mat.t().

    output image of the same size and the same number of channels as src.

The function applies an arbitrary linear filter to an image. In-place operation is supported. When the aperture is partially outside the image, the function interpolates outlier pixel values according to the specified border mode. The function does actually compute correlation, not the convolution: \f[\texttt{dst} (x,y) = \sum _{ \substack{0\leq x' < \texttt{kernel.cols}\\{0\leq y' < \texttt{kernel.rows}}}} \texttt{kernel} (x',y')* \texttt{src} (x+x'- \texttt{anchor.x} ,y+y'- \texttt{anchor.y} )\f] That is, the kernel is not mirrored around the anchor point. If you need a real convolution, flip the kernel using #flip and set the new anchor to (kernel.cols - anchor.x - 1, kernel.rows - anchor.y - 1). The function uses the DFT-based algorithm in case of sufficiently large kernels (~11 x 11 or larger) and the direct algorithm for small kernels. @sa sepFilter2D, dft, matchTemplate

Python prototype (for reference only):

filter2D(src, ddepth, kernel[, dst[, anchor[, delta[, borderType]]]]) -> dst
Link to this function

filter2D(src, ddepth, kernel, opts)

View Source
@spec filter2D(
  Evision.Mat.maybe_mat_in(),
  integer(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Convolves an image with the kernel.

Positional Arguments
  • src: Evision.Mat.t().

    input image.

  • ddepth: int.

    desired depth of the destination image, see @ref filter_depths "combinations"

  • kernel: Evision.Mat.t().

    convolution kernel (or rather a correlation kernel), a single-channel floating point matrix; if you want to apply different kernels to different channels, split the image into separate color planes using split and process them individually.

Keyword Arguments
  • anchor: Point.

    anchor of the kernel that indicates the relative position of a filtered point within the kernel; the anchor should lie within the kernel; default value (-1,-1) means that the anchor is at the kernel center.

  • delta: double.

    optional value added to the filtered pixels before storing them in dst.

  • borderType: int.

    pixel extrapolation method, see #BorderTypes. #BORDER_WRAP is not supported.

Return
  • dst: Evision.Mat.t().

    output image of the same size and the same number of channels as src.

The function applies an arbitrary linear filter to an image. In-place operation is supported. When the aperture is partially outside the image, the function interpolates outlier pixel values according to the specified border mode. The function does actually compute correlation, not the convolution: \f[\texttt{dst} (x,y) = \sum _{ \substack{0\leq x' < \texttt{kernel.cols}\\{0\leq y' < \texttt{kernel.rows}}}} \texttt{kernel} (x',y')* \texttt{src} (x+x'- \texttt{anchor.x} ,y+y'- \texttt{anchor.y} )\f] That is, the kernel is not mirrored around the anchor point. If you need a real convolution, flip the kernel using #flip and set the new anchor to (kernel.cols - anchor.x - 1, kernel.rows - anchor.y - 1). The function uses the DFT-based algorithm in case of sufficiently large kernels (~11 x 11 or larger) and the direct algorithm for small kernels. @sa sepFilter2D, dft, matchTemplate

Python prototype (for reference only):

filter2D(src, ddepth, kernel[, dst[, anchor[, delta[, borderType]]]]) -> dst
Link to this function

filterHomographyDecompByVisibleRefpoints(rotations, normals, beforePoints, afterPoints)

View Source
@spec filterHomographyDecompByVisibleRefpoints(
  [Evision.Mat.maybe_mat_in()],
  [Evision.Mat.maybe_mat_in()],
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in()
) :: Evision.Mat.t() | {:error, String.t()}

Filters homography decompositions based on additional information.

Positional Arguments
  • rotations: [Evision.Mat].

    Vector of rotation matrices.

  • normals: [Evision.Mat].

    Vector of plane normal matrices.

  • beforePoints: Evision.Mat.t().

    Vector of (rectified) visible reference points before the homography is applied

  • afterPoints: Evision.Mat.t().

    Vector of (rectified) visible reference points after the homography is applied

Keyword Arguments
  • pointsMask: Evision.Mat.t().

    optional Mat/Vector of 8u type representing the mask for the inliers as given by the #findHomography function

Return
  • possibleSolutions: Evision.Mat.t().

    Vector of int indices representing the viable solution set after filtering

This function is intended to filter the output of the #decomposeHomographyMat based on additional information as described in @cite Malis2007 . The summary of the method: the #decomposeHomographyMat function returns 2 unique solutions and their "opposites" for a total of 4 solutions. If we have access to the sets of points visible in the camera frame before and after the homography transformation is applied, we can determine which are the true potential solutions and which are the opposites by verifying which homographies are consistent with all visible reference points being in front of the camera. The inputs are left unchanged; the filtered solution set is returned as indices into the existing one.

Python prototype (for reference only):

filterHomographyDecompByVisibleRefpoints(rotations, normals, beforePoints, afterPoints[, possibleSolutions[, pointsMask]]) -> possibleSolutions
Link to this function

filterHomographyDecompByVisibleRefpoints(rotations, normals, beforePoints, afterPoints, opts)

View Source
@spec filterHomographyDecompByVisibleRefpoints(
  [Evision.Mat.maybe_mat_in()],
  [Evision.Mat.maybe_mat_in()],
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Filters homography decompositions based on additional information.

Positional Arguments
  • rotations: [Evision.Mat].

    Vector of rotation matrices.

  • normals: [Evision.Mat].

    Vector of plane normal matrices.

  • beforePoints: Evision.Mat.t().

    Vector of (rectified) visible reference points before the homography is applied

  • afterPoints: Evision.Mat.t().

    Vector of (rectified) visible reference points after the homography is applied

Keyword Arguments
  • pointsMask: Evision.Mat.t().

    optional Mat/Vector of 8u type representing the mask for the inliers as given by the #findHomography function

Return
  • possibleSolutions: Evision.Mat.t().

    Vector of int indices representing the viable solution set after filtering

This function is intended to filter the output of the #decomposeHomographyMat based on additional information as described in @cite Malis2007 . The summary of the method: the #decomposeHomographyMat function returns 2 unique solutions and their "opposites" for a total of 4 solutions. If we have access to the sets of points visible in the camera frame before and after the homography transformation is applied, we can determine which are the true potential solutions and which are the opposites by verifying which homographies are consistent with all visible reference points being in front of the camera. The inputs are left unchanged; the filtered solution set is returned as indices into the existing one.

Python prototype (for reference only):

filterHomographyDecompByVisibleRefpoints(rotations, normals, beforePoints, afterPoints[, possibleSolutions[, pointsMask]]) -> possibleSolutions
Link to this function

filterSpeckles(img, newVal, maxSpeckleSize, maxDiff)

View Source
@spec filterSpeckles(Evision.Mat.maybe_mat_in(), number(), integer(), number()) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Filters off small noise blobs (speckles) in the disparity map

Positional Arguments
  • newVal: double.

    The disparity value used to paint-off the speckles

  • maxSpeckleSize: int.

    The maximum speckle size to consider it a speckle. Larger blobs are not affected by the algorithm

  • maxDiff: double.

    Maximum difference between neighbor disparity pixels to put them into the same blob. Note that since StereoBM, StereoSGBM and may be other algorithms return a fixed-point disparity map, where disparity values are multiplied by 16, this scale factor should be taken into account when specifying this parameter value.

Return
  • img: Evision.Mat.t().

    The input 16-bit signed disparity image

  • buf: Evision.Mat.t().

    The optional temporary buffer to avoid memory allocation within the function.

Python prototype (for reference only):

filterSpeckles(img, newVal, maxSpeckleSize, maxDiff[, buf]) -> img, buf
Link to this function

filterSpeckles(img, newVal, maxSpeckleSize, maxDiff, opts)

View Source
@spec filterSpeckles(
  Evision.Mat.maybe_mat_in(),
  number(),
  integer(),
  number(),
  [{atom(), term()}, ...] | nil
) :: {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Filters off small noise blobs (speckles) in the disparity map

Positional Arguments
  • newVal: double.

    The disparity value used to paint-off the speckles

  • maxSpeckleSize: int.

    The maximum speckle size to consider it a speckle. Larger blobs are not affected by the algorithm

  • maxDiff: double.

    Maximum difference between neighbor disparity pixels to put them into the same blob. Note that since StereoBM, StereoSGBM and may be other algorithms return a fixed-point disparity map, where disparity values are multiplied by 16, this scale factor should be taken into account when specifying this parameter value.

Return
  • img: Evision.Mat.t().

    The input 16-bit signed disparity image

  • buf: Evision.Mat.t().

    The optional temporary buffer to avoid memory allocation within the function.

Python prototype (for reference only):

filterSpeckles(img, newVal, maxSpeckleSize, maxDiff[, buf]) -> img, buf
Link to this function

find4QuadCornerSubpix(img, corners, region_size)

View Source
@spec find4QuadCornerSubpix(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  {number(), number()}
) :: Evision.Mat.t() | false | {:error, String.t()}

find4QuadCornerSubpix

Positional Arguments
  • img: Evision.Mat.t()
  • region_size: Size
Return
  • retval: bool
  • corners: Evision.Mat.t()

Python prototype (for reference only):

find4QuadCornerSubpix(img, corners, region_size) -> retval, corners
Link to this function

findChessboardCorners(image, patternSize)

View Source
@spec findChessboardCorners(
  Evision.Mat.maybe_mat_in(),
  {number(), number()}
) :: Evision.Mat.t() | false | {:error, String.t()}

Finds the positions of internal corners of the chessboard.

Positional Arguments
  • image: Evision.Mat.t().

    Source chessboard view. It must be an 8-bit grayscale or color image.

  • patternSize: Size.

    Number of inner corners per a chessboard row and column ( patternSize = cv::Size(points_per_row,points_per_colum) = cv::Size(columns,rows) ).

Keyword Arguments
  • flags: int.Various operation flags that can be zero or a combination of the following values:
    • @ref CALIB_CB_ADAPTIVE_THRESH Use adaptive thresholding to convert the image to black and white, rather than a fixed threshold level (computed from the average image brightness).
    • @ref CALIB_CB_NORMALIZE_IMAGE Normalize the image gamma with #equalizeHist before applying fixed or adaptive thresholding.
    • @ref CALIB_CB_FILTER_QUADS Use additional criteria (like contour area, perimeter, square-like shape) to filter out false quads extracted at the contour retrieval stage.
    • @ref CALIB_CB_FAST_CHECK Run a fast check on the image that looks for chessboard corners, and shortcut the call if none is found. This can drastically speed up the call in the degenerate condition when no chessboard is observed.
    • @ref CALIB_CB_PLAIN All other flags are ignored. The input image is taken as is. No image processing is done to improve to find the checkerboard. This has the effect of speeding up the execution of the function but could lead to not recognizing the checkerboard if the image is not previously binarized in the appropriate manner.
Return
  • retval: bool

  • corners: Evision.Mat.t().

    Output array of detected corners.

The function attempts to determine whether the input image is a view of the chessboard pattern and locate the internal chessboard corners. The function returns a non-zero value if all of the corners are found and they are placed in a certain order (row by row, left to right in every row). Otherwise, if the function fails to find all the corners or reorder them, it returns 0. For example, a regular chessboard has 8 x 8 squares and 7 x 7 internal corners, that is, points where the black squares touch each other. The detected coordinates are approximate, and to determine their positions more accurately, the function calls #cornerSubPix. You also may use the function #cornerSubPix with different parameters if returned coordinates are not accurate enough. Sample usage of detecting and drawing chessboard corners: :

Size patternsize(8,6); //interior number of corners
Mat gray = ....; //source image
vector<Point2f> corners; //this will be filled by the detected corners
//CALIB_CB_FAST_CHECK saves a lot of time on images
//that do not contain any chessboard corners
bool patternfound = findChessboardCorners(gray, patternsize, corners,
CALIB_CB_ADAPTIVE_THRESH + CALIB_CB_NORMALIZE_IMAGE
+ CALIB_CB_FAST_CHECK);
if(patternfound)
cornerSubPix(gray, corners, Size(11, 11), Size(-1, -1),
TermCriteria(CV_TERMCRIT_EPS + CV_TERMCRIT_ITER, 30, 0.1));
drawChessboardCorners(img, patternsize, Mat(corners), patternfound);

Note: The function requires white space (like a square-thick border, the wider the better) around the board to make the detection more robust in various environments. Otherwise, if there is no border and the background is dark, the outer black squares cannot be segmented properly and so the square grouping and ordering algorithm fails. Use gen_pattern.py (@ref tutorial_camera_calibration_pattern) to create checkerboard.

Python prototype (for reference only):

findChessboardCorners(image, patternSize[, corners[, flags]]) -> retval, corners
Link to this function

findChessboardCorners(image, patternSize, opts)

View Source
@spec findChessboardCorners(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | false | {:error, String.t()}

Finds the positions of internal corners of the chessboard.

Positional Arguments
  • image: Evision.Mat.t().

    Source chessboard view. It must be an 8-bit grayscale or color image.

  • patternSize: Size.

    Number of inner corners per a chessboard row and column ( patternSize = cv::Size(points_per_row,points_per_colum) = cv::Size(columns,rows) ).

Keyword Arguments
  • flags: int.Various operation flags that can be zero or a combination of the following values:
    • @ref CALIB_CB_ADAPTIVE_THRESH Use adaptive thresholding to convert the image to black and white, rather than a fixed threshold level (computed from the average image brightness).
    • @ref CALIB_CB_NORMALIZE_IMAGE Normalize the image gamma with #equalizeHist before applying fixed or adaptive thresholding.
    • @ref CALIB_CB_FILTER_QUADS Use additional criteria (like contour area, perimeter, square-like shape) to filter out false quads extracted at the contour retrieval stage.
    • @ref CALIB_CB_FAST_CHECK Run a fast check on the image that looks for chessboard corners, and shortcut the call if none is found. This can drastically speed up the call in the degenerate condition when no chessboard is observed.
    • @ref CALIB_CB_PLAIN All other flags are ignored. The input image is taken as is. No image processing is done to improve to find the checkerboard. This has the effect of speeding up the execution of the function but could lead to not recognizing the checkerboard if the image is not previously binarized in the appropriate manner.
Return
  • retval: bool

  • corners: Evision.Mat.t().

    Output array of detected corners.

The function attempts to determine whether the input image is a view of the chessboard pattern and locate the internal chessboard corners. The function returns a non-zero value if all of the corners are found and they are placed in a certain order (row by row, left to right in every row). Otherwise, if the function fails to find all the corners or reorder them, it returns 0. For example, a regular chessboard has 8 x 8 squares and 7 x 7 internal corners, that is, points where the black squares touch each other. The detected coordinates are approximate, and to determine their positions more accurately, the function calls #cornerSubPix. You also may use the function #cornerSubPix with different parameters if returned coordinates are not accurate enough. Sample usage of detecting and drawing chessboard corners: :

Size patternsize(8,6); //interior number of corners
Mat gray = ....; //source image
vector<Point2f> corners; //this will be filled by the detected corners
//CALIB_CB_FAST_CHECK saves a lot of time on images
//that do not contain any chessboard corners
bool patternfound = findChessboardCorners(gray, patternsize, corners,
CALIB_CB_ADAPTIVE_THRESH + CALIB_CB_NORMALIZE_IMAGE
+ CALIB_CB_FAST_CHECK);
if(patternfound)
cornerSubPix(gray, corners, Size(11, 11), Size(-1, -1),
TermCriteria(CV_TERMCRIT_EPS + CV_TERMCRIT_ITER, 30, 0.1));
drawChessboardCorners(img, patternsize, Mat(corners), patternfound);

Note: The function requires white space (like a square-thick border, the wider the better) around the board to make the detection more robust in various environments. Otherwise, if there is no border and the background is dark, the outer black squares cannot be segmented properly and so the square grouping and ordering algorithm fails. Use gen_pattern.py (@ref tutorial_camera_calibration_pattern) to create checkerboard.

Python prototype (for reference only):

findChessboardCorners(image, patternSize[, corners[, flags]]) -> retval, corners
Link to this function

findChessboardCornersSB(image, patternSize)

View Source
@spec findChessboardCornersSB(
  Evision.Mat.maybe_mat_in(),
  {number(), number()}
) :: Evision.Mat.t() | false | {:error, String.t()}

findChessboardCornersSB

Positional Arguments
  • image: Evision.Mat.t()
  • patternSize: Size
Keyword Arguments
  • flags: int.
Return
  • retval: bool
  • corners: Evision.Mat.t().

Has overloading in C++

Python prototype (for reference only):

findChessboardCornersSB(image, patternSize[, corners[, flags]]) -> retval, corners
Link to this function

findChessboardCornersSB(image, patternSize, opts)

View Source
@spec findChessboardCornersSB(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | false | {:error, String.t()}

findChessboardCornersSB

Positional Arguments
  • image: Evision.Mat.t()
  • patternSize: Size
Keyword Arguments
  • flags: int.
Return
  • retval: bool
  • corners: Evision.Mat.t().

Has overloading in C++

Python prototype (for reference only):

findChessboardCornersSB(image, patternSize[, corners[, flags]]) -> retval, corners
Link to this function

findChessboardCornersSBWithMeta(image, patternSize, flags)

View Source
@spec findChessboardCornersSBWithMeta(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  integer()
) ::
  {Evision.Mat.t(), Evision.Mat.t()} | false | {:error, String.t()}

Finds the positions of internal corners of the chessboard using a sector based approach.

Positional Arguments
  • image: Evision.Mat.t().

    Source chessboard view. It must be an 8-bit grayscale or color image.

  • patternSize: Size.

    Number of inner corners per a chessboard row and column ( patternSize = cv::Size(points_per_row,points_per_colum) = cv::Size(columns,rows) ).

  • flags: int.

    Various operation flags that can be zero or a combination of the following values:

    • @ref CALIB_CB_NORMALIZE_IMAGE Normalize the image gamma with equalizeHist before detection.
    • @ref CALIB_CB_EXHAUSTIVE Run an exhaustive search to improve detection rate.
    • @ref CALIB_CB_ACCURACY Up sample input image to improve sub-pixel accuracy due to aliasing effects.
    • @ref CALIB_CB_LARGER The detected pattern is allowed to be larger than patternSize (see description).
    • @ref CALIB_CB_MARKER The detected pattern must have a marker (see description). This should be used if an accurate camera calibration is required.
Return
  • retval: bool

  • corners: Evision.Mat.t().

    Output array of detected corners.

  • meta: Evision.Mat.t().

    Optional output arrray of detected corners (CV_8UC1 and size = cv::Size(columns,rows)). Each entry stands for one corner of the pattern and can have one of the following values:

    • 0 = no meta data attached
    • 1 = left-top corner of a black cell
    • 2 = left-top corner of a white cell
    • 3 = left-top corner of a black cell with a white marker dot
    • 4 = left-top corner of a white cell with a black marker dot (pattern origin in case of markers otherwise first corner)

The function is analog to #findChessboardCorners but uses a localized radon transformation approximated by box filters being more robust to all sort of noise, faster on larger images and is able to directly return the sub-pixel position of the internal chessboard corners. The Method is based on the paper @cite duda2018 "Accurate Detection and Localization of Checkerboard Corners for Calibration" demonstrating that the returned sub-pixel positions are more accurate than the one returned by cornerSubPix allowing a precise camera calibration for demanding applications. In the case, the flags @ref CALIB_CB_LARGER or @ref CALIB_CB_MARKER are given, the result can be recovered from the optional meta array. Both flags are helpful to use calibration patterns exceeding the field of view of the camera. These oversized patterns allow more accurate calibrations as corners can be utilized, which are as close as possible to the image borders. For a consistent coordinate system across all images, the optional marker (see image below) can be used to move the origin of the board to the location where the black circle is located. Note: The function requires a white boarder with roughly the same width as one of the checkerboard fields around the whole board to improve the detection in various environments. In addition, because of the localized radon transformation it is beneficial to use round corners for the field corners which are located on the outside of the board. The following figure illustrates a sample checkerboard optimized for the detection. However, any other checkerboard can be used as well. Use gen_pattern.py (@ref tutorial_camera_calibration_pattern) to create checkerboard. Checkerboard

Python prototype (for reference only):

findChessboardCornersSBWithMeta(image, patternSize, flags[, corners[, meta]]) -> retval, corners, meta
Link to this function

findChessboardCornersSBWithMeta(image, patternSize, flags, opts)

View Source
@spec findChessboardCornersSBWithMeta(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  integer(),
  [{atom(), term()}, ...] | nil
) :: {Evision.Mat.t(), Evision.Mat.t()} | false | {:error, String.t()}

Finds the positions of internal corners of the chessboard using a sector based approach.

Positional Arguments
  • image: Evision.Mat.t().

    Source chessboard view. It must be an 8-bit grayscale or color image.

  • patternSize: Size.

    Number of inner corners per a chessboard row and column ( patternSize = cv::Size(points_per_row,points_per_colum) = cv::Size(columns,rows) ).

  • flags: int.

    Various operation flags that can be zero or a combination of the following values:

    • @ref CALIB_CB_NORMALIZE_IMAGE Normalize the image gamma with equalizeHist before detection.
    • @ref CALIB_CB_EXHAUSTIVE Run an exhaustive search to improve detection rate.
    • @ref CALIB_CB_ACCURACY Up sample input image to improve sub-pixel accuracy due to aliasing effects.
    • @ref CALIB_CB_LARGER The detected pattern is allowed to be larger than patternSize (see description).
    • @ref CALIB_CB_MARKER The detected pattern must have a marker (see description). This should be used if an accurate camera calibration is required.
Return
  • retval: bool

  • corners: Evision.Mat.t().

    Output array of detected corners.

  • meta: Evision.Mat.t().

    Optional output arrray of detected corners (CV_8UC1 and size = cv::Size(columns,rows)). Each entry stands for one corner of the pattern and can have one of the following values:

    • 0 = no meta data attached
    • 1 = left-top corner of a black cell
    • 2 = left-top corner of a white cell
    • 3 = left-top corner of a black cell with a white marker dot
    • 4 = left-top corner of a white cell with a black marker dot (pattern origin in case of markers otherwise first corner)

The function is analog to #findChessboardCorners but uses a localized radon transformation approximated by box filters being more robust to all sort of noise, faster on larger images and is able to directly return the sub-pixel position of the internal chessboard corners. The Method is based on the paper @cite duda2018 "Accurate Detection and Localization of Checkerboard Corners for Calibration" demonstrating that the returned sub-pixel positions are more accurate than the one returned by cornerSubPix allowing a precise camera calibration for demanding applications. In the case, the flags @ref CALIB_CB_LARGER or @ref CALIB_CB_MARKER are given, the result can be recovered from the optional meta array. Both flags are helpful to use calibration patterns exceeding the field of view of the camera. These oversized patterns allow more accurate calibrations as corners can be utilized, which are as close as possible to the image borders. For a consistent coordinate system across all images, the optional marker (see image below) can be used to move the origin of the board to the location where the black circle is located. Note: The function requires a white boarder with roughly the same width as one of the checkerboard fields around the whole board to improve the detection in various environments. In addition, because of the localized radon transformation it is beneficial to use round corners for the field corners which are located on the outside of the board. The following figure illustrates a sample checkerboard optimized for the detection. However, any other checkerboard can be used as well. Use gen_pattern.py (@ref tutorial_camera_calibration_pattern) to create checkerboard. Checkerboard

Python prototype (for reference only):

findChessboardCornersSBWithMeta(image, patternSize, flags[, corners[, meta]]) -> retval, corners, meta
Link to this function

findCirclesGrid(image, patternSize)

View Source
@spec findCirclesGrid(
  Evision.Mat.maybe_mat_in(),
  {number(), number()}
) :: Evision.Mat.t() | false | {:error, String.t()}

findCirclesGrid

Positional Arguments
  • image: Evision.Mat.t()
  • patternSize: Size
Keyword Arguments
  • flags: int.
  • blobDetector: FeatureDetector.
Return
  • retval: bool
  • centers: Evision.Mat.t().

Has overloading in C++

Python prototype (for reference only):

findCirclesGrid(image, patternSize[, centers[, flags[, blobDetector]]]) -> retval, centers
Link to this function

findCirclesGrid(image, patternSize, opts)

View Source
@spec findCirclesGrid(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | false | {:error, String.t()}

findCirclesGrid

Positional Arguments
  • image: Evision.Mat.t()
  • patternSize: Size
Keyword Arguments
  • flags: int.
  • blobDetector: FeatureDetector.
Return
  • retval: bool
  • centers: Evision.Mat.t().

Has overloading in C++

Python prototype (for reference only):

findCirclesGrid(image, patternSize[, centers[, flags[, blobDetector]]]) -> retval, centers
Link to this function

findCirclesGrid(image, patternSize, flags, blobDetector, parameters)

View Source
@spec findCirclesGrid(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  integer(),
  reference() | term(),
  Evision.CirclesGridFinderParameters.t()
) :: Evision.Mat.t() | false | {:error, String.t()}

Finds centers in the grid of circles.

Positional Arguments
  • image: Evision.Mat.t().

    grid view of input circles; it must be an 8-bit grayscale or color image.

  • patternSize: Size.

    number of circles per row and column ( patternSize = Size(points_per_row, points_per_colum) ).

  • flags: int.

    various operation flags that can be one of the following values:

    • @ref CALIB_CB_SYMMETRIC_GRID uses symmetric pattern of circles.
    • @ref CALIB_CB_ASYMMETRIC_GRID uses asymmetric pattern of circles.
    • @ref CALIB_CB_CLUSTERING uses a special algorithm for grid detection. It is more robust to perspective distortions but much more sensitive to background clutter.
  • blobDetector: FeatureDetector.

    feature detector that finds blobs like dark circles on light background. If blobDetector is NULL then image represents Point2f array of candidates.

  • parameters: Evision.CirclesGridFinderParameters.t().

    struct for finding circles in a grid pattern.

Return
  • retval: bool

  • centers: Evision.Mat.t().

    output array of detected centers.

The function attempts to determine whether the input image contains a grid of circles. If it is, the function locates centers of the circles. The function returns a non-zero value if all of the centers have been found and they have been placed in a certain order (row by row, left to right in every row). Otherwise, if the function fails to find all the corners or reorder them, it returns 0. Sample usage of detecting and drawing the centers of circles: :

Size patternsize(7,7); //number of centers
Mat gray = ...; //source image
vector<Point2f> centers; //this will be filled by the detected centers
bool patternfound = findCirclesGrid(gray, patternsize, centers);
drawChessboardCorners(img, patternsize, Mat(centers), patternfound);

Note: The function requires white space (like a square-thick border, the wider the better) around the board to make the detection more robust in various environments.

Python prototype (for reference only):

findCirclesGrid(image, patternSize, flags, blobDetector, parameters[, centers]) -> retval, centers
Link to this function

findCirclesGrid(image, patternSize, flags, blobDetector, parameters, opts)

View Source
@spec findCirclesGrid(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  integer(),
  reference() | term(),
  Evision.CirclesGridFinderParameters.t(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | false | {:error, String.t()}

Finds centers in the grid of circles.

Positional Arguments
  • image: Evision.Mat.t().

    grid view of input circles; it must be an 8-bit grayscale or color image.

  • patternSize: Size.

    number of circles per row and column ( patternSize = Size(points_per_row, points_per_colum) ).

  • flags: int.

    various operation flags that can be one of the following values:

    • @ref CALIB_CB_SYMMETRIC_GRID uses symmetric pattern of circles.
    • @ref CALIB_CB_ASYMMETRIC_GRID uses asymmetric pattern of circles.
    • @ref CALIB_CB_CLUSTERING uses a special algorithm for grid detection. It is more robust to perspective distortions but much more sensitive to background clutter.
  • blobDetector: FeatureDetector.

    feature detector that finds blobs like dark circles on light background. If blobDetector is NULL then image represents Point2f array of candidates.

  • parameters: Evision.CirclesGridFinderParameters.t().

    struct for finding circles in a grid pattern.

Return
  • retval: bool

  • centers: Evision.Mat.t().

    output array of detected centers.

The function attempts to determine whether the input image contains a grid of circles. If it is, the function locates centers of the circles. The function returns a non-zero value if all of the centers have been found and they have been placed in a certain order (row by row, left to right in every row). Otherwise, if the function fails to find all the corners or reorder them, it returns 0. Sample usage of detecting and drawing the centers of circles: :

Size patternsize(7,7); //number of centers
Mat gray = ...; //source image
vector<Point2f> centers; //this will be filled by the detected centers
bool patternfound = findCirclesGrid(gray, patternsize, centers);
drawChessboardCorners(img, patternsize, Mat(centers), patternfound);

Note: The function requires white space (like a square-thick border, the wider the better) around the board to make the detection more robust in various environments.

Python prototype (for reference only):

findCirclesGrid(image, patternSize, flags, blobDetector, parameters[, centers]) -> retval, centers
Link to this function

findContours(image, mode, method)

View Source
@spec findContours(Evision.Mat.maybe_mat_in(), integer(), integer()) ::
  {[Evision.Mat.t()], Evision.Mat.t()} | {:error, String.t()}

Finds contours in a binary image.

Positional Arguments
  • image: Evision.Mat.t().

    Source, an 8-bit single-channel image. Non-zero pixels are treated as 1's. Zero pixels remain 0's, so the image is treated as binary . You can use #compare, #inRange, #threshold , #adaptiveThreshold, #Canny, and others to create a binary image out of a grayscale or color one. If mode equals to #RETR_CCOMP or #RETR_FLOODFILL, the input can also be a 32-bit integer image of labels (CV_32SC1).

  • mode: int.

    Contour retrieval mode, see #RetrievalModes

  • method: int.

    Contour approximation method, see #ContourApproximationModes

Keyword Arguments
  • offset: Point.

    Optional offset by which every contour point is shifted. This is useful if the contours are extracted from the image ROI and then they should be analyzed in the whole image context.

Return
  • contours: [Evision.Mat].

    Detected contours. Each contour is stored as a vector of points (e.g. std::vector<std::vector<cv::Point> >).

  • hierarchy: Evision.Mat.t().

    Optional output vector (e.g. std::vector<cv::Vec4i>), containing information about the image topology. It has as many elements as the number of contours. For each i-th contour contours[i], the elements hierarchy[i][0] , hierarchy[i][1] , hierarchy[i][2] , and hierarchy[i][3] are set to 0-based indices in contours of the next and previous contours at the same hierarchical level, the first child contour and the parent contour, respectively. If for the contour i there are no next, previous, parent, or nested contours, the corresponding elements of hierarchy[i] will be negative.

The function retrieves contours from the binary image using the algorithm @cite Suzuki85 . The contours are a useful tool for shape analysis and object detection and recognition. See squares.cpp in the OpenCV sample directory. Note: Since opencv 3.2 source image is not modified by this function. Note: In Python, hierarchy is nested inside a top level array. Use hierarchy[0][i] to access hierarchical elements of i-th contour.

Python prototype (for reference only):

findContours(image, mode, method[, contours[, hierarchy[, offset]]]) -> contours, hierarchy
Link to this function

findContours(image, mode, method, opts)

View Source
@spec findContours(
  Evision.Mat.maybe_mat_in(),
  integer(),
  integer(),
  [{atom(), term()}, ...] | nil
) ::
  {[Evision.Mat.t()], Evision.Mat.t()} | {:error, String.t()}

Finds contours in a binary image.

Positional Arguments
  • image: Evision.Mat.t().

    Source, an 8-bit single-channel image. Non-zero pixels are treated as 1's. Zero pixels remain 0's, so the image is treated as binary . You can use #compare, #inRange, #threshold , #adaptiveThreshold, #Canny, and others to create a binary image out of a grayscale or color one. If mode equals to #RETR_CCOMP or #RETR_FLOODFILL, the input can also be a 32-bit integer image of labels (CV_32SC1).

  • mode: int.

    Contour retrieval mode, see #RetrievalModes

  • method: int.

    Contour approximation method, see #ContourApproximationModes

Keyword Arguments
  • offset: Point.

    Optional offset by which every contour point is shifted. This is useful if the contours are extracted from the image ROI and then they should be analyzed in the whole image context.

Return
  • contours: [Evision.Mat].

    Detected contours. Each contour is stored as a vector of points (e.g. std::vector<std::vector<cv::Point> >).

  • hierarchy: Evision.Mat.t().

    Optional output vector (e.g. std::vector<cv::Vec4i>), containing information about the image topology. It has as many elements as the number of contours. For each i-th contour contours[i], the elements hierarchy[i][0] , hierarchy[i][1] , hierarchy[i][2] , and hierarchy[i][3] are set to 0-based indices in contours of the next and previous contours at the same hierarchical level, the first child contour and the parent contour, respectively. If for the contour i there are no next, previous, parent, or nested contours, the corresponding elements of hierarchy[i] will be negative.

The function retrieves contours from the binary image using the algorithm @cite Suzuki85 . The contours are a useful tool for shape analysis and object detection and recognition. See squares.cpp in the OpenCV sample directory. Note: Since opencv 3.2 source image is not modified by this function. Note: In Python, hierarchy is nested inside a top level array. Use hierarchy[0][i] to access hierarchical elements of i-th contour.

Python prototype (for reference only):

findContours(image, mode, method[, contours[, hierarchy[, offset]]]) -> contours, hierarchy
Link to this function

findEssentialMat(points1, points2)

View Source
@spec findEssentialMat(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in()) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

findEssentialMat

Positional Arguments
  • points1: Evision.Mat.t().

    Array of N (N >= 5) 2D points from the first image. The point coordinates should be floating-point (single or double precision).

  • points2: Evision.Mat.t().

    Array of the second image points of the same size and format as points1 .

Keyword Arguments
  • focal: double.

    focal length of the camera. Note that this function assumes that points1 and points2 are feature points from cameras with same focal length and principal point.

  • pp: Point2d.

    principal point of the camera.

  • method: int.

    Method for computing a fundamental matrix.

    • @ref RANSAC for the RANSAC algorithm.
    • @ref LMEDS for the LMedS algorithm.
  • prob: double.

    Parameter used for the RANSAC or LMedS methods only. It specifies a desirable level of confidence (probability) that the estimated matrix is correct.

  • threshold: double.

    Parameter used for RANSAC. It is the maximum distance from a point to an epipolar line in pixels, beyond which the point is considered an outlier and is not used for computing the final fundamental matrix. It can be set to something like 1-3, depending on the accuracy of the point localization, image resolution, and the image noise.

  • maxIters: int.

    The maximum number of robust method iterations.

Return
  • retval: Evision.Mat.t()

  • mask: Evision.Mat.t().

    Output array of N elements, every element of which is set to 0 for outliers and to 1 for the other points. The array is computed only in the RANSAC and LMedS methods.

Has overloading in C++

This function differs from the one above that it computes camera intrinsic matrix from focal length and principal point: \f[A = \begin{bmatrix} f & 0 & x_{pp} \\ 0 & f & y_{pp} \\ 0 & 0 & 1 \end{bmatrix}\f]

Python prototype (for reference only):

findEssentialMat(points1, points2[, focal[, pp[, method[, prob[, threshold[, maxIters[, mask]]]]]]]) -> retval, mask
Link to this function

findEssentialMat(points1, points2, opts)

View Source
@spec findEssentialMat(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}
@spec findEssentialMat(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in()
) :: {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Variant 1:

Calculates an essential matrix from the corresponding points in two images.

Positional Arguments
  • points1: Evision.Mat.t().

    Array of N (N >= 5) 2D points from the first image. The point coordinates should be floating-point (single or double precision).

  • points2: Evision.Mat.t().

    Array of the second image points of the same size and format as points1 .

  • cameraMatrix: Evision.Mat.t().

    Camera intrinsic matrix \f$\cameramatrix{A}\f$ . Note that this function assumes that points1 and points2 are feature points from cameras with the same camera intrinsic matrix. If this assumption does not hold for your use case, use #undistortPoints with P = cv::NoArray() for both cameras to transform image points to normalized image coordinates, which are valid for the identity camera intrinsic matrix. When passing these coordinates, pass the identity matrix for this parameter.

Keyword Arguments
  • method: int.

    Method for computing an essential matrix.

    • @ref RANSAC for the RANSAC algorithm.
    • @ref LMEDS for the LMedS algorithm.
  • prob: double.

    Parameter used for the RANSAC or LMedS methods only. It specifies a desirable level of confidence (probability) that the estimated matrix is correct.

  • threshold: double.

    Parameter used for RANSAC. It is the maximum distance from a point to an epipolar line in pixels, beyond which the point is considered an outlier and is not used for computing the final fundamental matrix. It can be set to something like 1-3, depending on the accuracy of the point localization, image resolution, and the image noise.

  • maxIters: int.

    The maximum number of robust method iterations.

Return
  • retval: Evision.Mat.t()

  • mask: Evision.Mat.t().

    Output array of N elements, every element of which is set to 0 for outliers and to 1 for the other points. The array is computed only in the RANSAC and LMedS methods.

This function estimates essential matrix based on the five-point algorithm solver in @cite Nister03 . @cite SteweniusCFS is also a related. The epipolar geometry is described by the following equation: \f[[p_2; 1]^T K^{-T} E K^{-1} [p_1; 1] = 0\f] where \f$E\f$ is an essential matrix, \f$p_1\f$ and \f$p_2\f$ are corresponding points in the first and the second images, respectively. The result of this function may be passed further to #decomposeEssentialMat or #recoverPose to recover the relative pose between cameras.

Python prototype (for reference only):

findEssentialMat(points1, points2, cameraMatrix[, method[, prob[, threshold[, maxIters[, mask]]]]]) -> retval, mask

Variant 2:

findEssentialMat

Positional Arguments
  • points1: Evision.Mat.t().

    Array of N (N >= 5) 2D points from the first image. The point coordinates should be floating-point (single or double precision).

  • points2: Evision.Mat.t().

    Array of the second image points of the same size and format as points1 .

Keyword Arguments
  • focal: double.

    focal length of the camera. Note that this function assumes that points1 and points2 are feature points from cameras with same focal length and principal point.

  • pp: Point2d.

    principal point of the camera.

  • method: int.

    Method for computing a fundamental matrix.

    • @ref RANSAC for the RANSAC algorithm.
    • @ref LMEDS for the LMedS algorithm.
  • prob: double.

    Parameter used for the RANSAC or LMedS methods only. It specifies a desirable level of confidence (probability) that the estimated matrix is correct.

  • threshold: double.

    Parameter used for RANSAC. It is the maximum distance from a point to an epipolar line in pixels, beyond which the point is considered an outlier and is not used for computing the final fundamental matrix. It can be set to something like 1-3, depending on the accuracy of the point localization, image resolution, and the image noise.

  • maxIters: int.

    The maximum number of robust method iterations.

Return
  • retval: Evision.Mat.t()

  • mask: Evision.Mat.t().

    Output array of N elements, every element of which is set to 0 for outliers and to 1 for the other points. The array is computed only in the RANSAC and LMedS methods.

Has overloading in C++

This function differs from the one above that it computes camera intrinsic matrix from focal length and principal point: \f[A = \begin{bmatrix} f & 0 & x_{pp} \\ 0 & f & y_{pp} \\ 0 & 0 & 1 \end{bmatrix}\f]

Python prototype (for reference only):

findEssentialMat(points1, points2[, focal[, pp[, method[, prob[, threshold[, maxIters[, mask]]]]]]]) -> retval, mask
Link to this function

findEssentialMat(points1, points2, cameraMatrix, opts)

View Source
@spec findEssentialMat(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Calculates an essential matrix from the corresponding points in two images.

Positional Arguments
  • points1: Evision.Mat.t().

    Array of N (N >= 5) 2D points from the first image. The point coordinates should be floating-point (single or double precision).

  • points2: Evision.Mat.t().

    Array of the second image points of the same size and format as points1 .

  • cameraMatrix: Evision.Mat.t().

    Camera intrinsic matrix \f$\cameramatrix{A}\f$ . Note that this function assumes that points1 and points2 are feature points from cameras with the same camera intrinsic matrix. If this assumption does not hold for your use case, use #undistortPoints with P = cv::NoArray() for both cameras to transform image points to normalized image coordinates, which are valid for the identity camera intrinsic matrix. When passing these coordinates, pass the identity matrix for this parameter.

Keyword Arguments
  • method: int.

    Method for computing an essential matrix.

    • @ref RANSAC for the RANSAC algorithm.
    • @ref LMEDS for the LMedS algorithm.
  • prob: double.

    Parameter used for the RANSAC or LMedS methods only. It specifies a desirable level of confidence (probability) that the estimated matrix is correct.

  • threshold: double.

    Parameter used for RANSAC. It is the maximum distance from a point to an epipolar line in pixels, beyond which the point is considered an outlier and is not used for computing the final fundamental matrix. It can be set to something like 1-3, depending on the accuracy of the point localization, image resolution, and the image noise.

  • maxIters: int.

    The maximum number of robust method iterations.

Return
  • retval: Evision.Mat.t()

  • mask: Evision.Mat.t().

    Output array of N elements, every element of which is set to 0 for outliers and to 1 for the other points. The array is computed only in the RANSAC and LMedS methods.

This function estimates essential matrix based on the five-point algorithm solver in @cite Nister03 . @cite SteweniusCFS is also a related. The epipolar geometry is described by the following equation: \f[[p_2; 1]^T K^{-T} E K^{-1} [p_1; 1] = 0\f] where \f$E\f$ is an essential matrix, \f$p_1\f$ and \f$p_2\f$ are corresponding points in the first and the second images, respectively. The result of this function may be passed further to #decomposeEssentialMat or #recoverPose to recover the relative pose between cameras.

Python prototype (for reference only):

findEssentialMat(points1, points2, cameraMatrix[, method[, prob[, threshold[, maxIters[, mask]]]]]) -> retval, mask
Link to this function

findEssentialMat(points1, points2, cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2)

View Source

Calculates an essential matrix from the corresponding points in two images from potentially two different cameras.

Positional Arguments
  • points1: Evision.Mat.t().

    Array of N (N >= 5) 2D points from the first image. The point coordinates should be floating-point (single or double precision).

  • points2: Evision.Mat.t().

    Array of the second image points of the same size and format as points1 .

  • cameraMatrix1: Evision.Mat.t().

    Camera matrix \f$K = \vecthreethree{f_x}{0}{c_x}{0}{f_y}{c_y}{0}{0}{1}\f$ . Note that this function assumes that points1 and points2 are feature points from cameras with the same camera matrix. If this assumption does not hold for your use case, use #undistortPoints with P = cv::NoArray() for both cameras to transform image points to normalized image coordinates, which are valid for the identity camera matrix. When passing these coordinates, pass the identity matrix for this parameter.

  • distCoeffs1: Evision.Mat.t().

    Input vector of distortion coefficients \f$(k_1, k_2, p_1, p_2[, k_3[, k_4, k_5, k_6[, s_1, s_2, s_3, s_4[, \tau_x, \tau_y]]]])\f$ of 4, 5, 8, 12 or 14 elements. If the vector is NULL/empty, the zero distortion coefficients are assumed.

  • cameraMatrix2: Evision.Mat.t().

    Camera matrix \f$K = \vecthreethree{f_x}{0}{c_x}{0}{f_y}{c_y}{0}{0}{1}\f$ . Note that this function assumes that points1 and points2 are feature points from cameras with the same camera matrix. If this assumption does not hold for your use case, use #undistortPoints with P = cv::NoArray() for both cameras to transform image points to normalized image coordinates, which are valid for the identity camera matrix. When passing these coordinates, pass the identity matrix for this parameter.

  • distCoeffs2: Evision.Mat.t().

    Input vector of distortion coefficients \f$(k_1, k_2, p_1, p_2[, k_3[, k_4, k_5, k_6[, s_1, s_2, s_3, s_4[, \tau_x, \tau_y]]]])\f$ of 4, 5, 8, 12 or 14 elements. If the vector is NULL/empty, the zero distortion coefficients are assumed.

Keyword Arguments
  • method: int.

    Method for computing an essential matrix.

    • @ref RANSAC for the RANSAC algorithm.
    • @ref LMEDS for the LMedS algorithm.
  • prob: double.

    Parameter used for the RANSAC or LMedS methods only. It specifies a desirable level of confidence (probability) that the estimated matrix is correct.

  • threshold: double.

    Parameter used for RANSAC. It is the maximum distance from a point to an epipolar line in pixels, beyond which the point is considered an outlier and is not used for computing the final fundamental matrix. It can be set to something like 1-3, depending on the accuracy of the point localization, image resolution, and the image noise.

Return
  • retval: Evision.Mat.t()

  • mask: Evision.Mat.t().

    Output array of N elements, every element of which is set to 0 for outliers and to 1 for the other points. The array is computed only in the RANSAC and LMedS methods.

This function estimates essential matrix based on the five-point algorithm solver in @cite Nister03 . @cite SteweniusCFS is also a related. The epipolar geometry is described by the following equation: \f[[p_2; 1]^T K^{-T} E K^{-1} [p_1; 1] = 0\f] where \f$E\f$ is an essential matrix, \f$p_1\f$ and \f$p_2\f$ are corresponding points in the first and the second images, respectively. The result of this function may be passed further to #decomposeEssentialMat or #recoverPose to recover the relative pose between cameras.

Python prototype (for reference only):

findEssentialMat(points1, points2, cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2[, method[, prob[, threshold[, mask]]]]) -> retval, mask
Link to this function

findEssentialMat(points1, points2, cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2, opts)

View Source

Variant 1:

findEssentialMat

Positional Arguments
  • points1: Evision.Mat.t()
  • points2: Evision.Mat.t()
  • cameraMatrix1: Evision.Mat.t()
  • cameraMatrix2: Evision.Mat.t()
  • dist_coeff1: Evision.Mat.t()
  • dist_coeff2: Evision.Mat.t()
  • params: Evision.UsacParams.t()
Return
  • retval: Evision.Mat.t()
  • mask: Evision.Mat.t().

Python prototype (for reference only):

findEssentialMat(points1, points2, cameraMatrix1, cameraMatrix2, dist_coeff1, dist_coeff2, params[, mask]) -> retval, mask

Variant 2:

Calculates an essential matrix from the corresponding points in two images from potentially two different cameras.

Positional Arguments
  • points1: Evision.Mat.t().

    Array of N (N >= 5) 2D points from the first image. The point coordinates should be floating-point (single or double precision).

  • points2: Evision.Mat.t().

    Array of the second image points of the same size and format as points1 .

  • cameraMatrix1: Evision.Mat.t().

    Camera matrix \f$K = \vecthreethree{f_x}{0}{c_x}{0}{f_y}{c_y}{0}{0}{1}\f$ . Note that this function assumes that points1 and points2 are feature points from cameras with the same camera matrix. If this assumption does not hold for your use case, use #undistortPoints with P = cv::NoArray() for both cameras to transform image points to normalized image coordinates, which are valid for the identity camera matrix. When passing these coordinates, pass the identity matrix for this parameter.

  • distCoeffs1: Evision.Mat.t().

    Input vector of distortion coefficients \f$(k_1, k_2, p_1, p_2[, k_3[, k_4, k_5, k_6[, s_1, s_2, s_3, s_4[, \tau_x, \tau_y]]]])\f$ of 4, 5, 8, 12 or 14 elements. If the vector is NULL/empty, the zero distortion coefficients are assumed.

  • cameraMatrix2: Evision.Mat.t().

    Camera matrix \f$K = \vecthreethree{f_x}{0}{c_x}{0}{f_y}{c_y}{0}{0}{1}\f$ . Note that this function assumes that points1 and points2 are feature points from cameras with the same camera matrix. If this assumption does not hold for your use case, use #undistortPoints with P = cv::NoArray() for both cameras to transform image points to normalized image coordinates, which are valid for the identity camera matrix. When passing these coordinates, pass the identity matrix for this parameter.

  • distCoeffs2: Evision.Mat.t().

    Input vector of distortion coefficients \f$(k_1, k_2, p_1, p_2[, k_3[, k_4, k_5, k_6[, s_1, s_2, s_3, s_4[, \tau_x, \tau_y]]]])\f$ of 4, 5, 8, 12 or 14 elements. If the vector is NULL/empty, the zero distortion coefficients are assumed.

Keyword Arguments
  • method: int.

    Method for computing an essential matrix.

    • @ref RANSAC for the RANSAC algorithm.
    • @ref LMEDS for the LMedS algorithm.
  • prob: double.

    Parameter used for the RANSAC or LMedS methods only. It specifies a desirable level of confidence (probability) that the estimated matrix is correct.

  • threshold: double.

    Parameter used for RANSAC. It is the maximum distance from a point to an epipolar line in pixels, beyond which the point is considered an outlier and is not used for computing the final fundamental matrix. It can be set to something like 1-3, depending on the accuracy of the point localization, image resolution, and the image noise.

Return
  • retval: Evision.Mat.t()

  • mask: Evision.Mat.t().

    Output array of N elements, every element of which is set to 0 for outliers and to 1 for the other points. The array is computed only in the RANSAC and LMedS methods.

This function estimates essential matrix based on the five-point algorithm solver in @cite Nister03 . @cite SteweniusCFS is also a related. The epipolar geometry is described by the following equation: \f[[p_2; 1]^T K^{-T} E K^{-1} [p_1; 1] = 0\f] where \f$E\f$ is an essential matrix, \f$p_1\f$ and \f$p_2\f$ are corresponding points in the first and the second images, respectively. The result of this function may be passed further to #decomposeEssentialMat or #recoverPose to recover the relative pose between cameras.

Python prototype (for reference only):

findEssentialMat(points1, points2, cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2[, method[, prob[, threshold[, mask]]]]) -> retval, mask
Link to this function

findEssentialMat(points1, points2, cameraMatrix1, cameraMatrix2, dist_coeff1, dist_coeff2, params, opts)

View Source

findEssentialMat

Positional Arguments
  • points1: Evision.Mat.t()
  • points2: Evision.Mat.t()
  • cameraMatrix1: Evision.Mat.t()
  • cameraMatrix2: Evision.Mat.t()
  • dist_coeff1: Evision.Mat.t()
  • dist_coeff2: Evision.Mat.t()
  • params: Evision.UsacParams.t()
Return
  • retval: Evision.Mat.t()
  • mask: Evision.Mat.t().

Python prototype (for reference only):

findEssentialMat(points1, points2, cameraMatrix1, cameraMatrix2, dist_coeff1, dist_coeff2, params[, mask]) -> retval, mask
Link to this function

findFundamentalMat(points1, points2)

View Source
@spec findFundamentalMat(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in()) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

findFundamentalMat

Positional Arguments
  • points1: Evision.Mat.t()
  • points2: Evision.Mat.t()
Keyword Arguments
  • method: int.
  • ransacReprojThreshold: double.
  • confidence: double.
Return
  • retval: Evision.Mat.t()
  • mask: Evision.Mat.t().

Has overloading in C++

Python prototype (for reference only):

findFundamentalMat(points1, points2[, method[, ransacReprojThreshold[, confidence[, mask]]]]) -> retval, mask
Link to this function

findFundamentalMat(points1, points2, opts)

View Source
@spec findFundamentalMat(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}
@spec findFundamentalMat(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.UsacParams.t()
) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Variant 1:

findFundamentalMat

Positional Arguments
  • points1: Evision.Mat.t()
  • points2: Evision.Mat.t()
  • params: Evision.UsacParams.t()
Return
  • retval: Evision.Mat.t()
  • mask: Evision.Mat.t().

Has overloading in C++

Python prototype (for reference only):

findFundamentalMat(points1, points2, params[, mask]) -> retval, mask

Variant 2:

findFundamentalMat

Positional Arguments
  • points1: Evision.Mat.t()
  • points2: Evision.Mat.t()
Keyword Arguments
  • method: int.
  • ransacReprojThreshold: double.
  • confidence: double.
Return
  • retval: Evision.Mat.t()
  • mask: Evision.Mat.t().

Has overloading in C++

Python prototype (for reference only):

findFundamentalMat(points1, points2[, method[, ransacReprojThreshold[, confidence[, mask]]]]) -> retval, mask
Link to this function

findFundamentalMat(points1, points2, params, opts)

View Source
@spec findFundamentalMat(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.UsacParams.t(),
  [{atom(), term()}, ...] | nil
) :: {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

findFundamentalMat

Positional Arguments
  • points1: Evision.Mat.t()
  • points2: Evision.Mat.t()
  • params: Evision.UsacParams.t()
Return
  • retval: Evision.Mat.t()
  • mask: Evision.Mat.t().

Has overloading in C++

Python prototype (for reference only):

findFundamentalMat(points1, points2, params[, mask]) -> retval, mask
Link to this function

findFundamentalMat(points1, points2, method, ransacReprojThreshold, confidence, maxIters)

View Source
@spec findFundamentalMat(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  integer(),
  number(),
  number(),
  integer()
) :: {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Calculates a fundamental matrix from the corresponding points in two images.

Positional Arguments
  • points1: Evision.Mat.t().

    Array of N points from the first image. The point coordinates should be floating-point (single or double precision).

  • points2: Evision.Mat.t().

    Array of the second image points of the same size and format as points1 .

  • method: int.

    Method for computing a fundamental matrix.

    • @ref FM_7POINT for a 7-point algorithm. \f$N = 7\f$
    • @ref FM_8POINT for an 8-point algorithm. \f$N \ge 8\f$
    • @ref FM_RANSAC for the RANSAC algorithm. \f$N \ge 8\f$
    • @ref FM_LMEDS for the LMedS algorithm. \f$N \ge 8\f$
  • ransacReprojThreshold: double.

    Parameter used only for RANSAC. It is the maximum distance from a point to an epipolar line in pixels, beyond which the point is considered an outlier and is not used for computing the final fundamental matrix. It can be set to something like 1-3, depending on the accuracy of the point localization, image resolution, and the image noise.

  • confidence: double.

    Parameter used for the RANSAC and LMedS methods only. It specifies a desirable level of confidence (probability) that the estimated matrix is correct.

  • maxIters: int.

    The maximum number of robust method iterations.

Return
  • retval: Evision.Mat.t()

  • mask: Evision.Mat.t().

    optional output mask

The epipolar geometry is described by the following equation: \f[[p_2; 1]^T F [p_1; 1] = 0\f] where \f$F\f$ is a fundamental matrix, \f$p_1\f$ and \f$p_2\f$ are corresponding points in the first and the second images, respectively. The function calculates the fundamental matrix using one of four methods listed above and returns the found fundamental matrix. Normally just one matrix is found. But in case of the 7-point algorithm, the function may return up to 3 solutions ( \f$9 \times 3\f$ matrix that stores all 3 matrices sequentially). The calculated fundamental matrix may be passed further to #computeCorrespondEpilines that finds the epipolar lines corresponding to the specified points. It can also be passed to #stereoRectifyUncalibrated to compute the rectification transformation. :

// Example. Estimation of fundamental matrix using the RANSAC algorithm
int point_count = 100;
vector<Point2f> points1(point_count);
vector<Point2f> points2(point_count);
// initialize the points here ...
for( int i = 0; i < point_count; i++ )
{
points1[i] = ...;
points2[i] = ...;
}
Mat fundamental_matrix =
findFundamentalMat(points1, points2, FM_RANSAC, 3, 0.99);

Python prototype (for reference only):

findFundamentalMat(points1, points2, method, ransacReprojThreshold, confidence, maxIters[, mask]) -> retval, mask
Link to this function

findFundamentalMat(points1, points2, method, ransacReprojThreshold, confidence, maxIters, opts)

View Source
@spec findFundamentalMat(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  integer(),
  number(),
  number(),
  integer(),
  [{atom(), term()}, ...] | nil
) :: {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Calculates a fundamental matrix from the corresponding points in two images.

Positional Arguments
  • points1: Evision.Mat.t().

    Array of N points from the first image. The point coordinates should be floating-point (single or double precision).

  • points2: Evision.Mat.t().

    Array of the second image points of the same size and format as points1 .

  • method: int.

    Method for computing a fundamental matrix.

    • @ref FM_7POINT for a 7-point algorithm. \f$N = 7\f$
    • @ref FM_8POINT for an 8-point algorithm. \f$N \ge 8\f$
    • @ref FM_RANSAC for the RANSAC algorithm. \f$N \ge 8\f$
    • @ref FM_LMEDS for the LMedS algorithm. \f$N \ge 8\f$
  • ransacReprojThreshold: double.

    Parameter used only for RANSAC. It is the maximum distance from a point to an epipolar line in pixels, beyond which the point is considered an outlier and is not used for computing the final fundamental matrix. It can be set to something like 1-3, depending on the accuracy of the point localization, image resolution, and the image noise.

  • confidence: double.

    Parameter used for the RANSAC and LMedS methods only. It specifies a desirable level of confidence (probability) that the estimated matrix is correct.

  • maxIters: int.

    The maximum number of robust method iterations.

Return
  • retval: Evision.Mat.t()

  • mask: Evision.Mat.t().

    optional output mask

The epipolar geometry is described by the following equation: \f[[p_2; 1]^T F [p_1; 1] = 0\f] where \f$F\f$ is a fundamental matrix, \f$p_1\f$ and \f$p_2\f$ are corresponding points in the first and the second images, respectively. The function calculates the fundamental matrix using one of four methods listed above and returns the found fundamental matrix. Normally just one matrix is found. But in case of the 7-point algorithm, the function may return up to 3 solutions ( \f$9 \times 3\f$ matrix that stores all 3 matrices sequentially). The calculated fundamental matrix may be passed further to #computeCorrespondEpilines that finds the epipolar lines corresponding to the specified points. It can also be passed to #stereoRectifyUncalibrated to compute the rectification transformation. :

// Example. Estimation of fundamental matrix using the RANSAC algorithm
int point_count = 100;
vector<Point2f> points1(point_count);
vector<Point2f> points2(point_count);
// initialize the points here ...
for( int i = 0; i < point_count; i++ )
{
points1[i] = ...;
points2[i] = ...;
}
Mat fundamental_matrix =
findFundamentalMat(points1, points2, FM_RANSAC, 3, 0.99);

Python prototype (for reference only):

findFundamentalMat(points1, points2, method, ransacReprojThreshold, confidence, maxIters[, mask]) -> retval, mask
Link to this function

findHomography(srcPoints, dstPoints)

View Source
@spec findHomography(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in()) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Finds a perspective transformation between two planes.

Positional Arguments
  • srcPoints: Evision.Mat.t().

    Coordinates of the points in the original plane, a matrix of the type CV_32FC2 or vector\<Point2f> .

  • dstPoints: Evision.Mat.t().

    Coordinates of the points in the target plane, a matrix of the type CV_32FC2 or a vector\<Point2f> .

Keyword Arguments
  • method: int.

    Method used to compute a homography matrix. The following methods are possible:

    • 0 - a regular method using all the points, i.e., the least squares method
    • @ref RANSAC - RANSAC-based robust method
    • @ref LMEDS - Least-Median robust method
    • @ref RHO - PROSAC-based robust method
  • ransacReprojThreshold: double.

    Maximum allowed reprojection error to treat a point pair as an inlier (used in the RANSAC and RHO methods only). That is, if \f[\| \texttt{dstPoints} _i - \texttt{convertPointsHomogeneous} ( \texttt{H} \cdot \texttt{srcPoints} _i) \|_2 > \texttt{ransacReprojThreshold}\f] then the point \f$i\f$ is considered as an outlier. If srcPoints and dstPoints are measured in pixels, it usually makes sense to set this parameter somewhere in the range of 1 to 10.

  • maxIters: int.

    The maximum number of RANSAC iterations.

  • confidence: double.

    Confidence level, between 0 and 1.

Return
  • retval: Evision.Mat.t()

  • mask: Evision.Mat.t().

    Optional output mask set by a robust method ( RANSAC or LMeDS ). Note that the input mask values are ignored.

The function finds and returns the perspective transformation \f$H\f$ between the source and the destination planes: \f[s_i \vecthree{x'_i}{y'_i}{1} \sim H \vecthree{x_i}{y_i}{1}\f] so that the back-projection error \f[\sum _i \left ( x'_i- \frac{h_{11} x_i + h_{12} y_i + h_{13}}{h_{31} x_i + h_{32} y_i + h_{33}} \right )^2+ \left ( y'_i- \frac{h_{21} x_i + h_{22} y_i + h_{23}}{h_{31} x_i + h_{32} y_i + h_{33}} \right )^2\f] is minimized. If the parameter method is set to the default value 0, the function uses all the point pairs to compute an initial homography estimate with a simple least-squares scheme. However, if not all of the point pairs ( \f$srcPoints_i\f$, \f$dstPoints_i\f$ ) fit the rigid perspective transformation (that is, there are some outliers), this initial estimate will be poor. In this case, you can use one of the three robust methods. The methods RANSAC, LMeDS and RHO try many different random subsets of the corresponding point pairs (of four pairs each, collinear pairs are discarded), estimate the homography matrix using this subset and a simple least-squares algorithm, and then compute the quality/goodness of the computed homography (which is the number of inliers for RANSAC or the least median re-projection error for LMeDS). The best subset is then used to produce the initial estimate of the homography matrix and the mask of inliers/outliers. Regardless of the method, robust or not, the computed homography matrix is refined further (using inliers only in case of a robust method) with the Levenberg-Marquardt method to reduce the re-projection error even more. The methods RANSAC and RHO can handle practically any ratio of outliers but need a threshold to distinguish inliers from outliers. The method LMeDS does not need any threshold but it works correctly only when there are more than 50% of inliers. Finally, if there are no outliers and the noise is rather small, use the default method (method=0). The function is used to find initial intrinsic and extrinsic matrices. Homography matrix is determined up to a scale. Thus, it is normalized so that \f$h_{33}=1\f$. Note that whenever an \f$H\f$ matrix cannot be estimated, an empty one will be returned. @sa getAffineTransform, estimateAffine2D, estimateAffinePartial2D, getPerspectiveTransform, warpPerspective, perspectiveTransform

Python prototype (for reference only):

findHomography(srcPoints, dstPoints[, method[, ransacReprojThreshold[, mask[, maxIters[, confidence]]]]]) -> retval, mask
Link to this function

findHomography(srcPoints, dstPoints, opts)

View Source
@spec findHomography(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}
@spec findHomography(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.UsacParams.t()
) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Variant 1:

findHomography

Positional Arguments
  • srcPoints: Evision.Mat.t()
  • dstPoints: Evision.Mat.t()
  • params: Evision.UsacParams.t()
Return
  • retval: Evision.Mat.t()
  • mask: Evision.Mat.t().

Has overloading in C++

Python prototype (for reference only):

findHomography(srcPoints, dstPoints, params[, mask]) -> retval, mask

Variant 2:

Finds a perspective transformation between two planes.

Positional Arguments
  • srcPoints: Evision.Mat.t().

    Coordinates of the points in the original plane, a matrix of the type CV_32FC2 or vector\<Point2f> .

  • dstPoints: Evision.Mat.t().

    Coordinates of the points in the target plane, a matrix of the type CV_32FC2 or a vector\<Point2f> .

Keyword Arguments
  • method: int.

    Method used to compute a homography matrix. The following methods are possible:

    • 0 - a regular method using all the points, i.e., the least squares method
    • @ref RANSAC - RANSAC-based robust method
    • @ref LMEDS - Least-Median robust method
    • @ref RHO - PROSAC-based robust method
  • ransacReprojThreshold: double.

    Maximum allowed reprojection error to treat a point pair as an inlier (used in the RANSAC and RHO methods only). That is, if \f[\| \texttt{dstPoints} _i - \texttt{convertPointsHomogeneous} ( \texttt{H} \cdot \texttt{srcPoints} _i) \|_2 > \texttt{ransacReprojThreshold}\f] then the point \f$i\f$ is considered as an outlier. If srcPoints and dstPoints are measured in pixels, it usually makes sense to set this parameter somewhere in the range of 1 to 10.

  • maxIters: int.

    The maximum number of RANSAC iterations.

  • confidence: double.

    Confidence level, between 0 and 1.

Return
  • retval: Evision.Mat.t()

  • mask: Evision.Mat.t().

    Optional output mask set by a robust method ( RANSAC or LMeDS ). Note that the input mask values are ignored.

The function finds and returns the perspective transformation \f$H\f$ between the source and the destination planes: \f[s_i \vecthree{x'_i}{y'_i}{1} \sim H \vecthree{x_i}{y_i}{1}\f] so that the back-projection error \f[\sum _i \left ( x'_i- \frac{h_{11} x_i + h_{12} y_i + h_{13}}{h_{31} x_i + h_{32} y_i + h_{33}} \right )^2+ \left ( y'_i- \frac{h_{21} x_i + h_{22} y_i + h_{23}}{h_{31} x_i + h_{32} y_i + h_{33}} \right )^2\f] is minimized. If the parameter method is set to the default value 0, the function uses all the point pairs to compute an initial homography estimate with a simple least-squares scheme. However, if not all of the point pairs ( \f$srcPoints_i\f$, \f$dstPoints_i\f$ ) fit the rigid perspective transformation (that is, there are some outliers), this initial estimate will be poor. In this case, you can use one of the three robust methods. The methods RANSAC, LMeDS and RHO try many different random subsets of the corresponding point pairs (of four pairs each, collinear pairs are discarded), estimate the homography matrix using this subset and a simple least-squares algorithm, and then compute the quality/goodness of the computed homography (which is the number of inliers for RANSAC or the least median re-projection error for LMeDS). The best subset is then used to produce the initial estimate of the homography matrix and the mask of inliers/outliers. Regardless of the method, robust or not, the computed homography matrix is refined further (using inliers only in case of a robust method) with the Levenberg-Marquardt method to reduce the re-projection error even more. The methods RANSAC and RHO can handle practically any ratio of outliers but need a threshold to distinguish inliers from outliers. The method LMeDS does not need any threshold but it works correctly only when there are more than 50% of inliers. Finally, if there are no outliers and the noise is rather small, use the default method (method=0). The function is used to find initial intrinsic and extrinsic matrices. Homography matrix is determined up to a scale. Thus, it is normalized so that \f$h_{33}=1\f$. Note that whenever an \f$H\f$ matrix cannot be estimated, an empty one will be returned. @sa getAffineTransform, estimateAffine2D, estimateAffinePartial2D, getPerspectiveTransform, warpPerspective, perspectiveTransform

Python prototype (for reference only):

findHomography(srcPoints, dstPoints[, method[, ransacReprojThreshold[, mask[, maxIters[, confidence]]]]]) -> retval, mask
Link to this function

findHomography(srcPoints, dstPoints, params, opts)

View Source
@spec findHomography(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.UsacParams.t(),
  [{atom(), term()}, ...] | nil
) :: {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

findHomography

Positional Arguments
  • srcPoints: Evision.Mat.t()
  • dstPoints: Evision.Mat.t()
  • params: Evision.UsacParams.t()
Return
  • retval: Evision.Mat.t()
  • mask: Evision.Mat.t().

Has overloading in C++

Python prototype (for reference only):

findHomography(srcPoints, dstPoints, params[, mask]) -> retval, mask
@spec findNonZero(Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

Returns the list of locations of non-zero pixels

Positional Arguments
  • src: Evision.Mat.t().

    single-channel array

Return
  • idx: Evision.Mat.t().

    the output array, type of cv::Mat or std::vector<Point>, corresponding to non-zero indices in the input

Given a binary matrix (likely returned from an operation such as threshold(), compare(), >, ==, etc, return all of the non-zero indices as a cv::Mat or std::vector<cv::Point> (x,y) For example:

cv::Mat binaryImage; // input, binary image
cv::Mat locations;   // output, locations of non-zero pixels
cv::findNonZero(binaryImage, locations);
// access pixel coordinates
Point pnt = locations.at<Point>(i);

or

cv::Mat binaryImage; // input, binary image
vector<Point> locations;   // output, locations of non-zero pixels
cv::findNonZero(binaryImage, locations);
// access pixel coordinates
Point pnt = locations[i];

Python prototype (for reference only):

findNonZero(src[, idx]) -> idx
@spec findNonZero(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Returns the list of locations of non-zero pixels

Positional Arguments
  • src: Evision.Mat.t().

    single-channel array

Return
  • idx: Evision.Mat.t().

    the output array, type of cv::Mat or std::vector<Point>, corresponding to non-zero indices in the input

Given a binary matrix (likely returned from an operation such as threshold(), compare(), >, ==, etc, return all of the non-zero indices as a cv::Mat or std::vector<cv::Point> (x,y) For example:

cv::Mat binaryImage; // input, binary image
cv::Mat locations;   // output, locations of non-zero pixels
cv::findNonZero(binaryImage, locations);
// access pixel coordinates
Point pnt = locations.at<Point>(i);

or

cv::Mat binaryImage; // input, binary image
vector<Point> locations;   // output, locations of non-zero pixels
cv::findNonZero(binaryImage, locations);
// access pixel coordinates
Point pnt = locations[i];

Python prototype (for reference only):

findNonZero(src[, idx]) -> idx
Link to this function

findTransformECC(templateImage, inputImage, warpMatrix)

View Source

findTransformECC

Positional Arguments
  • templateImage: Evision.Mat.t()
  • inputImage: Evision.Mat.t()
Keyword Arguments
  • motionType: int.
  • criteria: TermCriteria.
  • inputMask: Evision.Mat.t().
Return
  • retval: double
  • warpMatrix: Evision.Mat.t()

Has overloading in C++

Python prototype (for reference only):

findTransformECC(templateImage, inputImage, warpMatrix[, motionType[, criteria[, inputMask]]]) -> retval, warpMatrix
Link to this function

findTransformECC(templateImage, inputImage, warpMatrix, opts)

View Source
@spec findTransformECC(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: {number(), Evision.Mat.t()} | {:error, String.t()}

findTransformECC

Positional Arguments
  • templateImage: Evision.Mat.t()
  • inputImage: Evision.Mat.t()
Keyword Arguments
  • motionType: int.
  • criteria: TermCriteria.
  • inputMask: Evision.Mat.t().
Return
  • retval: double
  • warpMatrix: Evision.Mat.t()

Has overloading in C++

Python prototype (for reference only):

findTransformECC(templateImage, inputImage, warpMatrix[, motionType[, criteria[, inputMask]]]) -> retval, warpMatrix
Link to this function

findTransformECC(templateImage, inputImage, warpMatrix, motionType, criteria, inputMask, gaussFiltSize)

View Source

Finds the geometric transform (warp) between two images in terms of the ECC criterion @cite EP08 .

Positional Arguments
  • templateImage: Evision.Mat.t().

    single-channel template image; CV_8U or CV_32F array.

  • inputImage: Evision.Mat.t().

    single-channel input image which should be warped with the final warpMatrix in order to provide an image similar to templateImage, same type as templateImage.

  • motionType: int.

    parameter, specifying the type of motion:

    • MOTION_TRANSLATION sets a translational motion model; warpMatrix is \f$2\times 3\f$ with the first \f$2\times 2\f$ part being the unity matrix and the rest two parameters being estimated.
    • MOTION_EUCLIDEAN sets a Euclidean (rigid) transformation as motion model; three parameters are estimated; warpMatrix is \f$2\times 3\f$.
    • MOTION_AFFINE sets an affine motion model (DEFAULT); six parameters are estimated; warpMatrix is \f$2\times 3\f$.
    • MOTION_HOMOGRAPHY sets a homography as a motion model; eight parameters are estimated;`warpMatrix` is \f$3\times 3\f$.
  • criteria: TermCriteria.

    parameter, specifying the termination criteria of the ECC algorithm; criteria.epsilon defines the threshold of the increment in the correlation coefficient between two iterations (a negative criteria.epsilon makes criteria.maxcount the only termination criterion). Default values are shown in the declaration above.

  • inputMask: Evision.Mat.t().

    An optional mask to indicate valid values of inputImage.

  • gaussFiltSize: int.

    An optional value indicating size of gaussian blur filter; (DEFAULT: 5)

Return
  • retval: double

  • warpMatrix: Evision.Mat.t().

    floating-point \f$2\times 3\f$ or \f$3\times 3\f$ mapping matrix (warp).

The function estimates the optimum transformation (warpMatrix) with respect to ECC criterion (@cite EP08), that is \f[\texttt{warpMatrix} = \arg\max_{W} \texttt{ECC}(\texttt{templateImage}(x,y),\texttt{inputImage}(x',y'))\f] where \f[\begin{bmatrix} x' \\ y' \end{bmatrix} = W \cdot \begin{bmatrix} x \\ y \\ 1 \end{bmatrix}\f] (the equation holds with homogeneous coordinates for homography). It returns the final enhanced correlation coefficient, that is the correlation coefficient between the template image and the final warped input image. When a \f$3\times 3\f$ matrix is given with motionType =0, 1 or 2, the third row is ignored. Unlike findHomography and estimateRigidTransform, the function findTransformECC implements an area-based alignment that builds on intensity similarities. In essence, the function updates the initial transformation that roughly aligns the images. If this information is missing, the identity warp (unity matrix) is used as an initialization. Note that if images undergo strong displacements/rotations, an initial transformation that roughly aligns the images is necessary (e.g., a simple euclidean/similarity transform that allows for the images showing the same image content approximately). Use inverse warping in the second image to take an image close to the first one, i.e. use the flag WARP_INVERSE_MAP with warpAffine or warpPerspective. See also the OpenCV sample image_alignment.cpp that demonstrates the use of the function. Note that the function throws an exception if algorithm does not converges. @sa computeECC, estimateAffine2D, estimateAffinePartial2D, findHomography

Python prototype (for reference only):

findTransformECC(templateImage, inputImage, warpMatrix, motionType, criteria, inputMask, gaussFiltSize) -> retval, warpMatrix
@spec fitEllipse(Evision.Mat.maybe_mat_in()) ::
  Evision.RotatedRect.t() | {:error, String.t()}

Fits an ellipse around a set of 2D points.

Positional Arguments
  • points: Evision.Mat.t().

    Input 2D point set, stored in std::vector\<> or Mat

Return
  • retval: {centre={x, y}, size={s1, s2}, angle}

The function calculates the ellipse that fits (in a least-squares sense) a set of 2D points best of all. It returns the rotated rectangle in which the ellipse is inscribed. The first algorithm described by @cite Fitzgibbon95 is used. Developer should keep in mind that it is possible that the returned ellipse/rotatedRect data contains negative indices, due to the data points being close to the border of the containing Mat element.

Python prototype (for reference only):

fitEllipse(points) -> retval
@spec fitEllipseAMS(Evision.Mat.maybe_mat_in()) ::
  Evision.RotatedRect.t() | {:error, String.t()}

Fits an ellipse around a set of 2D points.

Positional Arguments
  • points: Evision.Mat.t().

    Input 2D point set, stored in std::vector\<> or Mat

Return
  • retval: {centre={x, y}, size={s1, s2}, angle}

The function calculates the ellipse that fits a set of 2D points. It returns the rotated rectangle in which the ellipse is inscribed. The Approximate Mean Square (AMS) proposed by @cite Taubin1991 is used. For an ellipse, this basis set is \f$ \chi= \left(x^2, x y, y^2, x, y, 1\right) \f$, which is a set of six free coefficients \f$ A^T=\left\{A_{\text{xx}},A_{\text{xy}},A_{\text{yy}},A_x,A_y,A_0\right\} \f$. However, to specify an ellipse, all that is needed is five numbers; the major and minor axes lengths \f$ (a,b) \f$, the position \f$ (x_0,y_0) \f$, and the orientation \f$ \theta \f$. This is because the basis set includes lines, quadratics, parabolic and hyperbolic functions as well as elliptical functions as possible fits. If the fit is found to be a parabolic or hyperbolic function then the standard #fitEllipse method is used. The AMS method restricts the fit to parabolic, hyperbolic and elliptical curves by imposing the condition that \f$ A^T ( D_x^T D_x + D_y^T D_y) A = 1 \f$ where the matrices \f$ Dx \f$ and \f$ Dy \f$ are the partial derivatives of the design matrix \f$ D \f$ with respect to x and y. The matrices are formed row by row applying the following to each of the points in the set: \f{align}{ D(i,:)&=\left{x_i^2, x_i y_i, y_i^2, x_i, y_i, 1\right} & D_x(i,:)&=\left{2 x_i,y_i,0,1,0,0\right} & D_y(i,:)&=\left{0,x_i,2 y_i,0,1,0\right} \f} The AMS method minimizes the cost function \f{equation}{ \epsilon ^2=\frac{ A^T D^T D A }{ A^T (D_x^T D_x + D_y^T D_y) A^T } \f} The minimum cost is found by solving the generalized eigenvalue problem. \f{equation*}{ D^T D A = \lambda \left( D_x^T D_x + D_y^T D_y\right) A \f}

Python prototype (for reference only):

fitEllipseAMS(points) -> retval
Link to this function

fitEllipseDirect(points)

View Source
@spec fitEllipseDirect(Evision.Mat.maybe_mat_in()) ::
  Evision.RotatedRect.t() | {:error, String.t()}

Fits an ellipse around a set of 2D points.

Positional Arguments
  • points: Evision.Mat.t().

    Input 2D point set, stored in std::vector\<> or Mat

Return
  • retval: {centre={x, y}, size={s1, s2}, angle}

The function calculates the ellipse that fits a set of 2D points. It returns the rotated rectangle in which the ellipse is inscribed. The Direct least square (Direct) method by @cite Fitzgibbon1999 is used. For an ellipse, this basis set is \f$ \chi= \left(x^2, x y, y^2, x, y, 1\right) \f$, which is a set of six free coefficients \f$ A^T=\left\{A_{\text{xx}},A_{\text{xy}},A_{\text{yy}},A_x,A_y,A_0\right\} \f$. However, to specify an ellipse, all that is needed is five numbers; the major and minor axes lengths \f$ (a,b) \f$, the position \f$ (x_0,y_0) \f$, and the orientation \f$ \theta \f$. This is because the basis set includes lines, quadratics, parabolic and hyperbolic functions as well as elliptical functions as possible fits. The Direct method confines the fit to ellipses by ensuring that \f$ 4 A_{xx} A_{yy}- A_{xy}^2 > 0 \f$. The condition imposed is that \f$ 4 A_{xx} A_{yy}- A_{xy}^2=1 \f$ which satisfies the inequality and as the coefficients can be arbitrarily scaled is not overly restrictive. \f{equation}{ \epsilon ^2= A^T D^T D A \quad \text{with} \quad A^T C A =1 \quad \text{and} \quad C=\left(\begin{matrix} 0 & 0 & 2 & 0 & 0 & 0 \\ 0 & -1 & 0 & 0 & 0 & 0 \\ 2 & 0 & 0 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 & 0 & 0 \\ 0 & 0 & 0 & 0 & 0 & 0 \end{matrix} \right) \f} The minimum cost is found by solving the generalized eigenvalue problem. \f{equation}{ D^T D A = \lambda \left( C\right) A \f} The system produces only one positive eigenvalue \f$ \lambda\f$ which is chosen as the solution with its eigenvector \f$\mathbf{u}\f$. These are used to find the coefficients \f{equation*}{ A = \sqrt{\frac{1}{\mathbf{u}^T C \mathbf{u}}} \mathbf{u} \f} The scaling factor guarantees that \f$A^T C A =1\f$.

Python prototype (for reference only):

fitEllipseDirect(points) -> retval
Link to this function

fitLine(points, distType, param, reps, aeps)

View Source
@spec fitLine(Evision.Mat.maybe_mat_in(), integer(), number(), number(), number()) ::
  Evision.Mat.t() | {:error, String.t()}

Fits a line to a 2D or 3D point set.

Positional Arguments
  • points: Evision.Mat.t().

    Input vector of 2D or 3D points, stored in std::vector\<> or Mat.

  • distType: int.

    Distance used by the M-estimator, see #DistanceTypes

  • param: double.

    Numerical parameter ( C ) for some types of distances. If it is 0, an optimal value is chosen.

  • reps: double.

    Sufficient accuracy for the radius (distance between the coordinate origin and the line).

  • aeps: double.

    Sufficient accuracy for the angle. 0.01 would be a good default value for reps and aeps.

Return
  • line: Evision.Mat.t().

    Output line parameters. In case of 2D fitting, it should be a vector of 4 elements (like Vec4f) - (vx, vy, x0, y0), where (vx, vy) is a normalized vector collinear to the line and (x0, y0) is a point on the line. In case of 3D fitting, it should be a vector of 6 elements (like Vec6f) - (vx, vy, vz, x0, y0, z0), where (vx, vy, vz) is a normalized vector collinear to the line and (x0, y0, z0) is a point on the line.

The function fitLine fits a line to a 2D or 3D point set by minimizing \f$\sum_i \rho(r_i)\f$ where \f$r_i\f$ is a distance between the \f$i^{th}\f$ point, the line and \f$\rho(r)\f$ is a distance function, one of the following:

  • DIST_L2 \f[\rho (r) = r^2/2 \quad \text{(the simplest and the fastest least-squares method)}\f]

  • DIST_L1 \f[\rho (r) = r\f]

  • DIST_L12 \f[\rho (r) = 2 \cdot ( \sqrt{1 + \frac{r^2}{2}} - 1)\f]

  • DIST_FAIR \f[\rho \left (r \right ) = C^2 \cdot \left ( \frac{r}{C} - \log{\left(1 + \frac{r}{C}\right)} \right ) \quad \text{where} \quad C=1.3998\f]

  • DIST_WELSCH \f[\rho \left (r \right ) = \frac{C^2}{2} \cdot \left ( 1 - \exp{\left(-\left(\frac{r}{C}\right)^2\right)} \right ) \quad \text{where} \quad C=2.9846\f]

  • DIST_HUBER \f[\rho (r) = \fork{r^2/2}{if (r < C)}{C \cdot (r-C/2)}{otherwise} \quad \text{where} \quad C=1.345\f]

The algorithm is based on the M-estimator ( http://en.wikipedia.org/wiki/M-estimator ) technique that iteratively fits the line using the weighted least-squares algorithm. After each iteration the weights \f$w_i\f$ are adjusted to be inversely proportional to \f$\rho(r_i)\f$ .

Python prototype (for reference only):

fitLine(points, distType, param, reps, aeps[, line]) -> line
Link to this function

fitLine(points, distType, param, reps, aeps, opts)

View Source
@spec fitLine(
  Evision.Mat.maybe_mat_in(),
  integer(),
  number(),
  number(),
  number(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Fits a line to a 2D or 3D point set.

Positional Arguments
  • points: Evision.Mat.t().

    Input vector of 2D or 3D points, stored in std::vector\<> or Mat.

  • distType: int.

    Distance used by the M-estimator, see #DistanceTypes

  • param: double.

    Numerical parameter ( C ) for some types of distances. If it is 0, an optimal value is chosen.

  • reps: double.

    Sufficient accuracy for the radius (distance between the coordinate origin and the line).

  • aeps: double.

    Sufficient accuracy for the angle. 0.01 would be a good default value for reps and aeps.

Return
  • line: Evision.Mat.t().

    Output line parameters. In case of 2D fitting, it should be a vector of 4 elements (like Vec4f) - (vx, vy, x0, y0), where (vx, vy) is a normalized vector collinear to the line and (x0, y0) is a point on the line. In case of 3D fitting, it should be a vector of 6 elements (like Vec6f) - (vx, vy, vz, x0, y0, z0), where (vx, vy, vz) is a normalized vector collinear to the line and (x0, y0, z0) is a point on the line.

The function fitLine fits a line to a 2D or 3D point set by minimizing \f$\sum_i \rho(r_i)\f$ where \f$r_i\f$ is a distance between the \f$i^{th}\f$ point, the line and \f$\rho(r)\f$ is a distance function, one of the following:

  • DIST_L2 \f[\rho (r) = r^2/2 \quad \text{(the simplest and the fastest least-squares method)}\f]

  • DIST_L1 \f[\rho (r) = r\f]

  • DIST_L12 \f[\rho (r) = 2 \cdot ( \sqrt{1 + \frac{r^2}{2}} - 1)\f]

  • DIST_FAIR \f[\rho \left (r \right ) = C^2 \cdot \left ( \frac{r}{C} - \log{\left(1 + \frac{r}{C}\right)} \right ) \quad \text{where} \quad C=1.3998\f]

  • DIST_WELSCH \f[\rho \left (r \right ) = \frac{C^2}{2} \cdot \left ( 1 - \exp{\left(-\left(\frac{r}{C}\right)^2\right)} \right ) \quad \text{where} \quad C=2.9846\f]

  • DIST_HUBER \f[\rho (r) = \fork{r^2/2}{if (r < C)}{C \cdot (r-C/2)}{otherwise} \quad \text{where} \quad C=1.345\f]

The algorithm is based on the M-estimator ( http://en.wikipedia.org/wiki/M-estimator ) technique that iteratively fits the line using the weighted least-squares algorithm. After each iteration the weights \f$w_i\f$ are adjusted to be inversely proportional to \f$\rho(r_i)\f$ .

Python prototype (for reference only):

fitLine(points, distType, param, reps, aeps[, line]) -> line
@spec flip(Evision.Mat.maybe_mat_in(), integer()) ::
  Evision.Mat.t() | {:error, String.t()}

Flips a 2D array around vertical, horizontal, or both axes.

Positional Arguments
  • src: Evision.Mat.t().

    input array.

  • flipCode: int.

    a flag to specify how to flip the array; 0 means flipping around the x-axis and positive value (for example, 1) means flipping around y-axis. Negative value (for example, -1) means flipping around both axes.

Return
  • dst: Evision.Mat.t().

    output array of the same size and type as src.

The function cv::flip flips the array in one of three different ways (row and column indices are 0-based): \f[\texttt{dst} _{ij} = \left\{ \begin{array}{l l} \texttt{src} _{\texttt{src.rows}-i-1,j} & if\; \texttt{flipCode} = 0 \\ \texttt{src} _{i, \texttt{src.cols} -j-1} & if\; \texttt{flipCode} > 0 \\ \texttt{src} _{ \texttt{src.rows} -i-1, \texttt{src.cols} -j-1} & if\; \texttt{flipCode} < 0 \\ \end{array} \right.\f] The example scenarios of using the function are the following: Vertical flipping of the image (flipCode == 0) to switch between top-left and bottom-left image origin. This is a typical operation in video processing on Microsoft Windows* OS. Horizontal flipping of the image with the subsequent horizontal shift and absolute difference calculation to check for a vertical-axis symmetry (flipCode > 0). Simultaneous horizontal and vertical flipping of the image with the subsequent shift and absolute difference calculation to check for a central symmetry (flipCode \< 0). Reversing the order of point arrays (flipCode > 0 or flipCode == 0). @sa transpose , repeat , completeSymm

Python prototype (for reference only):

flip(src, flipCode[, dst]) -> dst
Link to this function

flip(src, flipCode, opts)

View Source
@spec flip(Evision.Mat.maybe_mat_in(), integer(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Flips a 2D array around vertical, horizontal, or both axes.

Positional Arguments
  • src: Evision.Mat.t().

    input array.

  • flipCode: int.

    a flag to specify how to flip the array; 0 means flipping around the x-axis and positive value (for example, 1) means flipping around y-axis. Negative value (for example, -1) means flipping around both axes.

Return
  • dst: Evision.Mat.t().

    output array of the same size and type as src.

The function cv::flip flips the array in one of three different ways (row and column indices are 0-based): \f[\texttt{dst} _{ij} = \left\{ \begin{array}{l l} \texttt{src} _{\texttt{src.rows}-i-1,j} & if\; \texttt{flipCode} = 0 \\ \texttt{src} _{i, \texttt{src.cols} -j-1} & if\; \texttt{flipCode} > 0 \\ \texttt{src} _{ \texttt{src.rows} -i-1, \texttt{src.cols} -j-1} & if\; \texttt{flipCode} < 0 \\ \end{array} \right.\f] The example scenarios of using the function are the following: Vertical flipping of the image (flipCode == 0) to switch between top-left and bottom-left image origin. This is a typical operation in video processing on Microsoft Windows* OS. Horizontal flipping of the image with the subsequent horizontal shift and absolute difference calculation to check for a vertical-axis symmetry (flipCode > 0). Simultaneous horizontal and vertical flipping of the image with the subsequent shift and absolute difference calculation to check for a central symmetry (flipCode \< 0). Reversing the order of point arrays (flipCode > 0 or flipCode == 0). @sa transpose , repeat , completeSymm

Python prototype (for reference only):

flip(src, flipCode[, dst]) -> dst
@spec flipND(Evision.Mat.maybe_mat_in(), integer()) ::
  Evision.Mat.t() | {:error, String.t()}

Flips a n-dimensional at given axis

Positional Arguments
  • src: Evision.Mat.t().

    input array

  • axis: int.

    axis that performs a flip on. 0 <= axis < src.dims.

Return
  • dst: Evision.Mat.t().

    output array that has the same shape of src

Python prototype (for reference only):

flipND(src, axis[, dst]) -> dst
@spec flipND(Evision.Mat.maybe_mat_in(), integer(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Flips a n-dimensional at given axis

Positional Arguments
  • src: Evision.Mat.t().

    input array

  • axis: int.

    axis that performs a flip on. 0 <= axis < src.dims.

Return
  • dst: Evision.Mat.t().

    output array that has the same shape of src

Python prototype (for reference only):

flipND(src, axis[, dst]) -> dst
Link to this function

floodFill(image, mask, seedPoint, newVal)

View Source
@spec floodFill(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()}
) ::
  {integer(), Evision.Mat.t(), Evision.Mat.t(),
   {number(), number(), number(), number()}}
  | {:error, String.t()}

Fills a connected component with the given color.

Positional Arguments
  • seedPoint: Point.

    Starting point.

  • newVal: Scalar.

    New value of the repainted domain pixels.

Keyword Arguments
  • loDiff: Scalar.

    Maximal lower brightness/color difference between the currently observed pixel and one of its neighbors belonging to the component, or a seed pixel being added to the component.

  • upDiff: Scalar.

    Maximal upper brightness/color difference between the currently observed pixel and one of its neighbors belonging to the component, or a seed pixel being added to the component.

  • flags: int.

    Operation flags. The first 8 bits contain a connectivity value. The default value of 4 means that only the four nearest neighbor pixels (those that share an edge) are considered. A connectivity value of 8 means that the eight nearest neighbor pixels (those that share a corner) will be considered. The next 8 bits (8-16) contain a value between 1 and 255 with which to fill the mask (the default value is 1). For example, 4 | ( 255 \<\< 8 ) will consider 4 nearest neighbours and fill the mask with a value of 255. The following additional options occupy higher bits and therefore may be further combined with the connectivity and mask fill values using bit-wise or (|), see #FloodFillFlags.

Return
  • retval: int

  • image: Evision.Mat.t().

    Input/output 1- or 3-channel, 8-bit, or floating-point image. It is modified by the function unless the #FLOODFILL_MASK_ONLY flag is set in the second variant of the function. See the details below.

  • mask: Evision.Mat.t().

    Operation mask that should be a single-channel 8-bit image, 2 pixels wider and 2 pixels taller than image. If an empty Mat is passed it will be created automatically. Since this is both an input and output parameter, you must take responsibility of initializing it. Flood-filling cannot go across non-zero pixels in the input mask. For example, an edge detector output can be used as a mask to stop filling at edges. On output, pixels in the mask corresponding to filled pixels in the image are set to 1 or to the specified value in flags as described below. Additionally, the function fills the border of the mask with ones to simplify internal processing. It is therefore possible to use the same mask in multiple calls to the function to make sure the filled areas do not overlap.

  • rect: Rect*.

    Optional output parameter set by the function to the minimum bounding rectangle of the repainted domain.

The function cv::floodFill fills a connected component starting from the seed point with the specified color. The connectivity is determined by the color/brightness closeness of the neighbor pixels. The pixel at \f$(x,y)\f$ is considered to belong to the repainted domain if:

  • in case of a grayscale image and floating range \f[\texttt{src} (x',y')- \texttt{loDiff} \leq \texttt{src} (x,y) \leq \texttt{src} (x',y')+ \texttt{upDiff}\f]

  • in case of a grayscale image and fixed range \f[\texttt{src} ( \texttt{seedPoint} .x, \texttt{seedPoint} .y)- \texttt{loDiff} \leq \texttt{src} (x,y) \leq \texttt{src} ( \texttt{seedPoint} .x, \texttt{seedPoint} .y)+ \texttt{upDiff}\f]

  • in case of a color image and floating range \f[\texttt{src} (x',y')_r- \texttt{loDiff} _r \leq \texttt{src} (x,y)_r \leq \texttt{src} (x',y')_r+ \texttt{upDiff} _r,\f] \f[\texttt{src} (x',y')_g- \texttt{loDiff} _g \leq \texttt{src} (x,y)_g \leq \texttt{src} (x',y')_g+ \texttt{upDiff} _g\f] and \f[\texttt{src} (x',y')_b- \texttt{loDiff} _b \leq \texttt{src} (x,y)_b \leq \texttt{src} (x',y')_b+ \texttt{upDiff} _b\f]

  • in case of a color image and fixed range \f[\texttt{src} ( \texttt{seedPoint} .x, \texttt{seedPoint} .y)_r- \texttt{loDiff} _r \leq \texttt{src} (x,y)_r \leq \texttt{src} ( \texttt{seedPoint} .x, \texttt{seedPoint} .y)_r+ \texttt{upDiff} _r,\f] \f[\texttt{src} ( \texttt{seedPoint} .x, \texttt{seedPoint} .y)_g- \texttt{loDiff} _g \leq \texttt{src} (x,y)_g \leq \texttt{src} ( \texttt{seedPoint} .x, \texttt{seedPoint} .y)_g+ \texttt{upDiff} _g\f] and \f[\texttt{src} ( \texttt{seedPoint} .x, \texttt{seedPoint} .y)_b- \texttt{loDiff} _b \leq \texttt{src} (x,y)_b \leq \texttt{src} ( \texttt{seedPoint} .x, \texttt{seedPoint} .y)_b+ \texttt{upDiff} _b\f]

where \f$src(x',y')\f$ is the value of one of pixel neighbors that is already known to belong to the component. That is, to be added to the connected component, a color/brightness of the pixel should be close enough to:

  • Color/brightness of one of its neighbors that already belong to the connected component in case of a floating range.

  • Color/brightness of the seed point in case of a fixed range.

Use these functions to either mark a connected component with the specified color in-place, or build a mask and then extract the contour, or copy the region to another image, and so on.

Note: Since the mask is larger than the filled image, a pixel \f$(x, y)\f$ in image corresponds to the pixel \f$(x+1, y+1)\f$ in the mask . @sa findContours

Python prototype (for reference only):

floodFill(image, mask, seedPoint, newVal[, loDiff[, upDiff[, flags]]]) -> retval, image, mask, rect
Link to this function

floodFill(image, mask, seedPoint, newVal, opts)

View Source
@spec floodFill(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()},
  [{atom(), term()}, ...] | nil
) ::
  {integer(), Evision.Mat.t(), Evision.Mat.t(),
   {number(), number(), number(), number()}}
  | {:error, String.t()}

Fills a connected component with the given color.

Positional Arguments
  • seedPoint: Point.

    Starting point.

  • newVal: Scalar.

    New value of the repainted domain pixels.

Keyword Arguments
  • loDiff: Scalar.

    Maximal lower brightness/color difference between the currently observed pixel and one of its neighbors belonging to the component, or a seed pixel being added to the component.

  • upDiff: Scalar.

    Maximal upper brightness/color difference between the currently observed pixel and one of its neighbors belonging to the component, or a seed pixel being added to the component.

  • flags: int.

    Operation flags. The first 8 bits contain a connectivity value. The default value of 4 means that only the four nearest neighbor pixels (those that share an edge) are considered. A connectivity value of 8 means that the eight nearest neighbor pixels (those that share a corner) will be considered. The next 8 bits (8-16) contain a value between 1 and 255 with which to fill the mask (the default value is 1). For example, 4 | ( 255 \<\< 8 ) will consider 4 nearest neighbours and fill the mask with a value of 255. The following additional options occupy higher bits and therefore may be further combined with the connectivity and mask fill values using bit-wise or (|), see #FloodFillFlags.

Return
  • retval: int

  • image: Evision.Mat.t().

    Input/output 1- or 3-channel, 8-bit, or floating-point image. It is modified by the function unless the #FLOODFILL_MASK_ONLY flag is set in the second variant of the function. See the details below.

  • mask: Evision.Mat.t().

    Operation mask that should be a single-channel 8-bit image, 2 pixels wider and 2 pixels taller than image. If an empty Mat is passed it will be created automatically. Since this is both an input and output parameter, you must take responsibility of initializing it. Flood-filling cannot go across non-zero pixels in the input mask. For example, an edge detector output can be used as a mask to stop filling at edges. On output, pixels in the mask corresponding to filled pixels in the image are set to 1 or to the specified value in flags as described below. Additionally, the function fills the border of the mask with ones to simplify internal processing. It is therefore possible to use the same mask in multiple calls to the function to make sure the filled areas do not overlap.

  • rect: Rect*.

    Optional output parameter set by the function to the minimum bounding rectangle of the repainted domain.

The function cv::floodFill fills a connected component starting from the seed point with the specified color. The connectivity is determined by the color/brightness closeness of the neighbor pixels. The pixel at \f$(x,y)\f$ is considered to belong to the repainted domain if:

  • in case of a grayscale image and floating range \f[\texttt{src} (x',y')- \texttt{loDiff} \leq \texttt{src} (x,y) \leq \texttt{src} (x',y')+ \texttt{upDiff}\f]

  • in case of a grayscale image and fixed range \f[\texttt{src} ( \texttt{seedPoint} .x, \texttt{seedPoint} .y)- \texttt{loDiff} \leq \texttt{src} (x,y) \leq \texttt{src} ( \texttt{seedPoint} .x, \texttt{seedPoint} .y)+ \texttt{upDiff}\f]

  • in case of a color image and floating range \f[\texttt{src} (x',y')_r- \texttt{loDiff} _r \leq \texttt{src} (x,y)_r \leq \texttt{src} (x',y')_r+ \texttt{upDiff} _r,\f] \f[\texttt{src} (x',y')_g- \texttt{loDiff} _g \leq \texttt{src} (x,y)_g \leq \texttt{src} (x',y')_g+ \texttt{upDiff} _g\f] and \f[\texttt{src} (x',y')_b- \texttt{loDiff} _b \leq \texttt{src} (x,y)_b \leq \texttt{src} (x',y')_b+ \texttt{upDiff} _b\f]

  • in case of a color image and fixed range \f[\texttt{src} ( \texttt{seedPoint} .x, \texttt{seedPoint} .y)_r- \texttt{loDiff} _r \leq \texttt{src} (x,y)_r \leq \texttt{src} ( \texttt{seedPoint} .x, \texttt{seedPoint} .y)_r+ \texttt{upDiff} _r,\f] \f[\texttt{src} ( \texttt{seedPoint} .x, \texttt{seedPoint} .y)_g- \texttt{loDiff} _g \leq \texttt{src} (x,y)_g \leq \texttt{src} ( \texttt{seedPoint} .x, \texttt{seedPoint} .y)_g+ \texttt{upDiff} _g\f] and \f[\texttt{src} ( \texttt{seedPoint} .x, \texttt{seedPoint} .y)_b- \texttt{loDiff} _b \leq \texttt{src} (x,y)_b \leq \texttt{src} ( \texttt{seedPoint} .x, \texttt{seedPoint} .y)_b+ \texttt{upDiff} _b\f]

where \f$src(x',y')\f$ is the value of one of pixel neighbors that is already known to belong to the component. That is, to be added to the connected component, a color/brightness of the pixel should be close enough to:

  • Color/brightness of one of its neighbors that already belong to the connected component in case of a floating range.

  • Color/brightness of the seed point in case of a fixed range.

Use these functions to either mark a connected component with the specified color in-place, or build a mask and then extract the contour, or copy the region to another image, and so on.

Note: Since the mask is larger than the filled image, a pixel \f$(x, y)\f$ in image corresponds to the pixel \f$(x+1, y+1)\f$ in the mask . @sa findContours

Python prototype (for reference only):

floodFill(image, mask, seedPoint, newVal[, loDiff[, upDiff[, flags]]]) -> retval, image, mask, rect
Link to this function

gaussianBlur(src, ksize, sigmaX)

View Source
@spec gaussianBlur(Evision.Mat.maybe_mat_in(), {number(), number()}, number()) ::
  Evision.Mat.t() | {:error, String.t()}

Blurs an image using a Gaussian filter.

Positional Arguments
  • src: Evision.Mat.t().

    input image; the image can have any number of channels, which are processed independently, but the depth should be CV_8U, CV_16U, CV_16S, CV_32F or CV_64F.

  • ksize: Size.

    Gaussian kernel size. ksize.width and ksize.height can differ but they both must be positive and odd. Or, they can be zero's and then they are computed from sigma.

  • sigmaX: double.

    Gaussian kernel standard deviation in X direction.

Keyword Arguments
  • sigmaY: double.

    Gaussian kernel standard deviation in Y direction; if sigmaY is zero, it is set to be equal to sigmaX, if both sigmas are zeros, they are computed from ksize.width and ksize.height, respectively (see #getGaussianKernel for details); to fully control the result regardless of possible future modifications of all this semantics, it is recommended to specify all of ksize, sigmaX, and sigmaY.

  • borderType: int.

    pixel extrapolation method, see #BorderTypes. #BORDER_WRAP is not supported.

Return
  • dst: Evision.Mat.t().

    output image of the same size and type as src.

The function convolves the source image with the specified Gaussian kernel. In-place filtering is supported.

@sa sepFilter2D, filter2D, blur, boxFilter, bilateralFilter, medianBlur

Python prototype (for reference only):

GaussianBlur(src, ksize, sigmaX[, dst[, sigmaY[, borderType]]]) -> dst
Link to this function

gaussianBlur(src, ksize, sigmaX, opts)

View Source
@spec gaussianBlur(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  number(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Blurs an image using a Gaussian filter.

Positional Arguments
  • src: Evision.Mat.t().

    input image; the image can have any number of channels, which are processed independently, but the depth should be CV_8U, CV_16U, CV_16S, CV_32F or CV_64F.

  • ksize: Size.

    Gaussian kernel size. ksize.width and ksize.height can differ but they both must be positive and odd. Or, they can be zero's and then they are computed from sigma.

  • sigmaX: double.

    Gaussian kernel standard deviation in X direction.

Keyword Arguments
  • sigmaY: double.

    Gaussian kernel standard deviation in Y direction; if sigmaY is zero, it is set to be equal to sigmaX, if both sigmas are zeros, they are computed from ksize.width and ksize.height, respectively (see #getGaussianKernel for details); to fully control the result regardless of possible future modifications of all this semantics, it is recommended to specify all of ksize, sigmaX, and sigmaY.

  • borderType: int.

    pixel extrapolation method, see #BorderTypes. #BORDER_WRAP is not supported.

Return
  • dst: Evision.Mat.t().

    output image of the same size and type as src.

The function convolves the source image with the specified Gaussian kernel. In-place filtering is supported.

@sa sepFilter2D, filter2D, blur, boxFilter, bilateralFilter, medianBlur

Python prototype (for reference only):

GaussianBlur(src, ksize, sigmaX[, dst[, sigmaY[, borderType]]]) -> dst
Link to this function

gemm(src1, src2, alpha, src3, beta)

View Source

Performs generalized matrix multiplication.

Positional Arguments
  • src1: Evision.Mat.t().

    first multiplied input matrix that could be real(CV_32FC1, CV_64FC1) or complex(CV_32FC2, CV_64FC2).

  • src2: Evision.Mat.t().

    second multiplied input matrix of the same type as src1.

  • alpha: double.

    weight of the matrix product.

  • src3: Evision.Mat.t().

    third optional delta matrix added to the matrix product; it should have the same type as src1 and src2.

  • beta: double.

    weight of src3.

Keyword Arguments
  • flags: int.

    operation flags (cv::GemmFlags)

Return
  • dst: Evision.Mat.t().

    output matrix; it has the proper size and the same type as input matrices.

The function cv::gemm performs generalized matrix multiplication similar to the gemm functions in BLAS level 3. For example, gemm(src1, src2, alpha, src3, beta, dst, GEMM_1_T + GEMM_3_T) corresponds to \f[\texttt{dst} = \texttt{alpha} \cdot \texttt{src1} ^T \cdot \texttt{src2} + \texttt{beta} \cdot \texttt{src3} ^T\f] In case of complex (two-channel) data, performed a complex matrix multiplication. The function can be replaced with a matrix expression. For example, the above call can be replaced with:

dst = alpha*src1.t()*src2 + beta*src3.t();

@sa mulTransposed , transform

Python prototype (for reference only):

gemm(src1, src2, alpha, src3, beta[, dst[, flags]]) -> dst
Link to this function

gemm(src1, src2, alpha, src3, beta, opts)

View Source
@spec gemm(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  number(),
  Evision.Mat.maybe_mat_in(),
  number(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Performs generalized matrix multiplication.

Positional Arguments
  • src1: Evision.Mat.t().

    first multiplied input matrix that could be real(CV_32FC1, CV_64FC1) or complex(CV_32FC2, CV_64FC2).

  • src2: Evision.Mat.t().

    second multiplied input matrix of the same type as src1.

  • alpha: double.

    weight of the matrix product.

  • src3: Evision.Mat.t().

    third optional delta matrix added to the matrix product; it should have the same type as src1 and src2.

  • beta: double.

    weight of src3.

Keyword Arguments
  • flags: int.

    operation flags (cv::GemmFlags)

Return
  • dst: Evision.Mat.t().

    output matrix; it has the proper size and the same type as input matrices.

The function cv::gemm performs generalized matrix multiplication similar to the gemm functions in BLAS level 3. For example, gemm(src1, src2, alpha, src3, beta, dst, GEMM_1_T + GEMM_3_T) corresponds to \f[\texttt{dst} = \texttt{alpha} \cdot \texttt{src1} ^T \cdot \texttt{src2} + \texttt{beta} \cdot \texttt{src3} ^T\f] In case of complex (two-channel) data, performed a complex matrix multiplication. The function can be replaced with a matrix expression. For example, the above call can be replaced with:

dst = alpha*src1.t()*src2 + beta*src3.t();

@sa mulTransposed , transform

Python prototype (for reference only):

gemm(src1, src2, alpha, src3, beta[, dst[, flags]]) -> dst
Link to this function

getAffineTransform(src, dst)

View Source
@spec getAffineTransform(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

getAffineTransform

Positional Arguments
  • src: Evision.Mat.t()
  • dst: Evision.Mat.t()
Return
  • retval: Evision.Mat.t()

Has overloading in C++

Python prototype (for reference only):

getAffineTransform(src, dst) -> retval
@spec getBuildInformation() :: binary() | {:error, String.t()}

Returns full configuration time cmake output.

Return

Returned value is raw cmake output including version control system revision, compiler version, compiler flags, enabled modules and third party libraries, etc. Output format depends on target architecture.

Python prototype (for reference only):

getBuildInformation() -> retval
@spec getCPUFeaturesLine() :: binary() | {:error, String.t()}

Returns list of CPU features enabled during compilation.

Return
  • retval: string

Returned value is a string containing space separated list of CPU features with following markers:

  • no markers - baseline features
  • prefix * - features enabled in dispatcher
  • suffix ? - features enabled but not available in HW

Example: SSE SSE2 SSE3 *SSE4.1 *SSE4.2 *FP16 *AVX *AVX2 *AVX512-SKX?

Python prototype (for reference only):

getCPUFeaturesLine() -> retval
@spec getCPUTickCount() :: integer() | {:error, String.t()}

Returns the number of CPU ticks.

Return
  • retval: int64

The function returns the current number of CPU ticks on some architectures (such as x86, x64, PowerPC). On other platforms the function is equivalent to getTickCount. It can also be used for very accurate time measurements, as well as for RNG initialization. Note that in case of multi-CPU systems a thread, from which getCPUTickCount is called, can be suspended and resumed at another CPU with its own counter. So, theoretically (and practically) the subsequent calls to the function do not necessary return the monotonously increasing values. Also, since a modern CPU varies the CPU frequency depending on the load, the number of CPU clocks spent in some code cannot be directly converted to time units. Therefore, getTickCount is generally a preferable solution for measuring execution time.

Python prototype (for reference only):

getCPUTickCount() -> retval
Link to this function

getDefaultNewCameraMatrix(cameraMatrix)

View Source
@spec getDefaultNewCameraMatrix(Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

Returns the default new camera matrix.

Positional Arguments
  • cameraMatrix: Evision.Mat.t().

    Input camera matrix.

Keyword Arguments
  • imgsize: Size.

    Camera view image size in pixels.

  • centerPrincipalPoint: bool.

    Location of the principal point in the new camera matrix. The parameter indicates whether this location should be at the image center or not.

Return
  • retval: Evision.Mat.t()

The function returns the camera matrix that is either an exact copy of the input cameraMatrix (when centerPrinicipalPoint=false ), or the modified one (when centerPrincipalPoint=true). In the latter case, the new camera matrix will be: \f[\begin{bmatrix} f_x && 0 && ( \texttt{imgSize.width} -1)*0.5 \\ 0 && f_y && ( \texttt{imgSize.height} -1)*0.5 \\ 0 && 0 && 1 \end{bmatrix} ,\f] where \f$f_x\f$ and \f$f_y\f$ are \f$(0,0)\f$ and \f$(1,1)\f$ elements of cameraMatrix, respectively. By default, the undistortion functions in OpenCV (see #initUndistortRectifyMap, #undistort) do not move the principal point. However, when you work with stereo, it is important to move the principal points in both views to the same y-coordinate (which is required by most of stereo correspondence algorithms), and may be to the same x-coordinate too. So, you can form the new camera matrix for each view where the principal points are located at the center.

Python prototype (for reference only):

getDefaultNewCameraMatrix(cameraMatrix[, imgsize[, centerPrincipalPoint]]) -> retval
Link to this function

getDefaultNewCameraMatrix(cameraMatrix, opts)

View Source
@spec getDefaultNewCameraMatrix(
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Returns the default new camera matrix.

Positional Arguments
  • cameraMatrix: Evision.Mat.t().

    Input camera matrix.

Keyword Arguments
  • imgsize: Size.

    Camera view image size in pixels.

  • centerPrincipalPoint: bool.

    Location of the principal point in the new camera matrix. The parameter indicates whether this location should be at the image center or not.

Return
  • retval: Evision.Mat.t()

The function returns the camera matrix that is either an exact copy of the input cameraMatrix (when centerPrinicipalPoint=false ), or the modified one (when centerPrincipalPoint=true). In the latter case, the new camera matrix will be: \f[\begin{bmatrix} f_x && 0 && ( \texttt{imgSize.width} -1)*0.5 \\ 0 && f_y && ( \texttt{imgSize.height} -1)*0.5 \\ 0 && 0 && 1 \end{bmatrix} ,\f] where \f$f_x\f$ and \f$f_y\f$ are \f$(0,0)\f$ and \f$(1,1)\f$ elements of cameraMatrix, respectively. By default, the undistortion functions in OpenCV (see #initUndistortRectifyMap, #undistort) do not move the principal point. However, when you work with stereo, it is important to move the principal points in both views to the same y-coordinate (which is required by most of stereo correspondence algorithms), and may be to the same x-coordinate too. So, you can form the new camera matrix for each view where the principal points are located at the center.

Python prototype (for reference only):

getDefaultNewCameraMatrix(cameraMatrix[, imgsize[, centerPrincipalPoint]]) -> retval
Link to this function

getDerivKernels(dx, dy, ksize)

View Source
@spec getDerivKernels(integer(), integer(), integer()) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Returns filter coefficients for computing spatial image derivatives.

Positional Arguments
  • dx: int.

    Derivative order in respect of x.

  • dy: int.

    Derivative order in respect of y.

  • ksize: int.

    Aperture size. It can be FILTER_SCHARR, 1, 3, 5, or 7.

Keyword Arguments
  • normalize: bool.

    Flag indicating whether to normalize (scale down) the filter coefficients or not. Theoretically, the coefficients should have the denominator \f$=2^{ksize*2-dx-dy-2}\f$. If you are going to filter floating-point images, you are likely to use the normalized kernels. But if you compute derivatives of an 8-bit image, store the results in a 16-bit image, and wish to preserve all the fractional bits, you may want to set normalize=false .

  • ktype: int.

    Type of filter coefficients. It can be CV_32f or CV_64F .

Return
  • kx: Evision.Mat.t().

    Output matrix of row filter coefficients. It has the type ktype .

  • ky: Evision.Mat.t().

    Output matrix of column filter coefficients. It has the type ktype .

The function computes and returns the filter coefficients for spatial image derivatives. When ksize=FILTER_SCHARR, the Scharr \f$3 \times 3\f$ kernels are generated (see #Scharr). Otherwise, Sobel kernels are generated (see #Sobel). The filters are normally passed to #sepFilter2D or to

Python prototype (for reference only):

getDerivKernels(dx, dy, ksize[, kx[, ky[, normalize[, ktype]]]]) -> kx, ky
Link to this function

getDerivKernels(dx, dy, ksize, opts)

View Source
@spec getDerivKernels(integer(), integer(), integer(), [{atom(), term()}, ...] | nil) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Returns filter coefficients for computing spatial image derivatives.

Positional Arguments
  • dx: int.

    Derivative order in respect of x.

  • dy: int.

    Derivative order in respect of y.

  • ksize: int.

    Aperture size. It can be FILTER_SCHARR, 1, 3, 5, or 7.

Keyword Arguments
  • normalize: bool.

    Flag indicating whether to normalize (scale down) the filter coefficients or not. Theoretically, the coefficients should have the denominator \f$=2^{ksize*2-dx-dy-2}\f$. If you are going to filter floating-point images, you are likely to use the normalized kernels. But if you compute derivatives of an 8-bit image, store the results in a 16-bit image, and wish to preserve all the fractional bits, you may want to set normalize=false .

  • ktype: int.

    Type of filter coefficients. It can be CV_32f or CV_64F .

Return
  • kx: Evision.Mat.t().

    Output matrix of row filter coefficients. It has the type ktype .

  • ky: Evision.Mat.t().

    Output matrix of column filter coefficients. It has the type ktype .

The function computes and returns the filter coefficients for spatial image derivatives. When ksize=FILTER_SCHARR, the Scharr \f$3 \times 3\f$ kernels are generated (see #Scharr). Otherwise, Sobel kernels are generated (see #Sobel). The filters are normally passed to #sepFilter2D or to

Python prototype (for reference only):

getDerivKernels(dx, dy, ksize[, kx[, ky[, normalize[, ktype]]]]) -> kx, ky
Link to this function

getFontScaleFromHeight(fontFace, pixelHeight)

View Source
@spec getFontScaleFromHeight(integer(), integer()) :: number() | {:error, String.t()}

Calculates the font-specific size to use to achieve a given height in pixels.

Positional Arguments
  • fontFace: int.

    Font to use, see cv::HersheyFonts.

  • pixelHeight: int.

    Pixel height to compute the fontScale for

Keyword Arguments
  • thickness: int.

    Thickness of lines used to render the text.See putText for details.

Return
  • retval: double

@return The fontSize to use for cv::putText @see cv::putText

Python prototype (for reference only):

getFontScaleFromHeight(fontFace, pixelHeight[, thickness]) -> retval
Link to this function

getFontScaleFromHeight(fontFace, pixelHeight, opts)

View Source
@spec getFontScaleFromHeight(integer(), integer(), [{atom(), term()}, ...] | nil) ::
  number() | {:error, String.t()}

Calculates the font-specific size to use to achieve a given height in pixels.

Positional Arguments
  • fontFace: int.

    Font to use, see cv::HersheyFonts.

  • pixelHeight: int.

    Pixel height to compute the fontScale for

Keyword Arguments
  • thickness: int.

    Thickness of lines used to render the text.See putText for details.

Return
  • retval: double

@return The fontSize to use for cv::putText @see cv::putText

Python prototype (for reference only):

getFontScaleFromHeight(fontFace, pixelHeight[, thickness]) -> retval
Link to this function

getGaborKernel(ksize, sigma, theta, lambd, gamma)

View Source
@spec getGaborKernel({number(), number()}, number(), number(), number(), number()) ::
  Evision.Mat.t() | {:error, String.t()}

Returns Gabor filter coefficients.

Positional Arguments
  • ksize: Size.

    Size of the filter returned.

  • sigma: double.

    Standard deviation of the gaussian envelope.

  • theta: double.

    Orientation of the normal to the parallel stripes of a Gabor function.

  • lambd: double.

    Wavelength of the sinusoidal factor.

  • gamma: double.

    Spatial aspect ratio.

Keyword Arguments
  • psi: double.

    Phase offset.

  • ktype: int.

    Type of filter coefficients. It can be CV_32F or CV_64F .

Return
  • retval: Evision.Mat.t()

For more details about gabor filter equations and parameters, see: Gabor Filter.

Python prototype (for reference only):

getGaborKernel(ksize, sigma, theta, lambd, gamma[, psi[, ktype]]) -> retval
Link to this function

getGaborKernel(ksize, sigma, theta, lambd, gamma, opts)

View Source
@spec getGaborKernel(
  {number(), number()},
  number(),
  number(),
  number(),
  number(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Returns Gabor filter coefficients.

Positional Arguments
  • ksize: Size.

    Size of the filter returned.

  • sigma: double.

    Standard deviation of the gaussian envelope.

  • theta: double.

    Orientation of the normal to the parallel stripes of a Gabor function.

  • lambd: double.

    Wavelength of the sinusoidal factor.

  • gamma: double.

    Spatial aspect ratio.

Keyword Arguments
  • psi: double.

    Phase offset.

  • ktype: int.

    Type of filter coefficients. It can be CV_32F or CV_64F .

Return
  • retval: Evision.Mat.t()

For more details about gabor filter equations and parameters, see: Gabor Filter.

Python prototype (for reference only):

getGaborKernel(ksize, sigma, theta, lambd, gamma[, psi[, ktype]]) -> retval
Link to this function

getGaussianKernel(ksize, sigma)

View Source
@spec getGaussianKernel(integer(), number()) :: Evision.Mat.t() | {:error, String.t()}

Returns Gaussian filter coefficients.

Positional Arguments
  • ksize: int.

    Aperture size. It should be odd ( \f$\texttt{ksize} \mod 2 = 1\f$ ) and positive.

  • sigma: double.

    Gaussian standard deviation. If it is non-positive, it is computed from ksize as sigma = 0.3*((ksize-1)*0.5 - 1) + 0.8.

Keyword Arguments
  • ktype: int.

    Type of filter coefficients. It can be CV_32F or CV_64F .

Return
  • retval: Evision.Mat.t()

The function computes and returns the \f$\texttt{ksize} \times 1\f$ matrix of Gaussian filter coefficients: \f[G_i= \alpha *e^{-(i-( \texttt{ksize} -1)/2)^2/(2* \texttt{sigma}^2)},\f] where \f$i=0..\texttt{ksize}-1\f$ and \f$\alpha\f$ is the scale factor chosen so that \f$\sum_i G_i=1\f$. Two of such generated kernels can be passed to sepFilter2D. Those functions automatically recognize smoothing kernels (a symmetrical kernel with sum of weights equal to 1) and handle them accordingly. You may also use the higher-level GaussianBlur. @sa sepFilter2D, getDerivKernels, getStructuringElement, GaussianBlur

Python prototype (for reference only):

getGaussianKernel(ksize, sigma[, ktype]) -> retval
Link to this function

getGaussianKernel(ksize, sigma, opts)

View Source
@spec getGaussianKernel(integer(), number(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Returns Gaussian filter coefficients.

Positional Arguments
  • ksize: int.

    Aperture size. It should be odd ( \f$\texttt{ksize} \mod 2 = 1\f$ ) and positive.

  • sigma: double.

    Gaussian standard deviation. If it is non-positive, it is computed from ksize as sigma = 0.3*((ksize-1)*0.5 - 1) + 0.8.

Keyword Arguments
  • ktype: int.

    Type of filter coefficients. It can be CV_32F or CV_64F .

Return
  • retval: Evision.Mat.t()

The function computes and returns the \f$\texttt{ksize} \times 1\f$ matrix of Gaussian filter coefficients: \f[G_i= \alpha *e^{-(i-( \texttt{ksize} -1)/2)^2/(2* \texttt{sigma}^2)},\f] where \f$i=0..\texttt{ksize}-1\f$ and \f$\alpha\f$ is the scale factor chosen so that \f$\sum_i G_i=1\f$. Two of such generated kernels can be passed to sepFilter2D. Those functions automatically recognize smoothing kernels (a symmetrical kernel with sum of weights equal to 1) and handle them accordingly. You may also use the higher-level GaussianBlur. @sa sepFilter2D, getDerivKernels, getStructuringElement, GaussianBlur

Python prototype (for reference only):

getGaussianKernel(ksize, sigma[, ktype]) -> retval
Link to this function

getHardwareFeatureName(feature)

View Source
@spec getHardwareFeatureName(integer()) :: binary() | {:error, String.t()}

Returns feature name by ID

Positional Arguments
  • feature: int
Return

Returns empty string if feature is not defined

Python prototype (for reference only):

getHardwareFeatureName(feature) -> retval
@spec getLogLevel() :: integer() | {:error, String.t()}

getLogLevel

Return
  • retval: int

Python prototype (for reference only):

getLogLevel() -> retval
@spec getNumberOfCPUs() :: integer() | {:error, String.t()}

Returns the number of logical CPUs available for the process.

Return
  • retval: int

Python prototype (for reference only):

getNumberOfCPUs() -> retval
@spec getNumThreads() :: integer() | {:error, String.t()}

Returns the number of threads used by OpenCV for parallel regions.

Return
  • retval: int

Always returns 1 if OpenCV is built without threading support. The exact meaning of return value depends on the threading framework used by OpenCV library:

  • TBB - The number of threads, that OpenCV will try to use for parallel regions. If there is any tbb::thread_scheduler_init in user code conflicting with OpenCV, then function returns default number of threads used by TBB library.

  • OpenMP - An upper bound on the number of threads that could be used to form a new team.

  • Concurrency - The number of threads, that OpenCV will try to use for parallel regions.

  • GCD - Unsupported; returns the GCD thread pool limit (512) for compatibility.

  • C= - The number of threads, that OpenCV will try to use for parallel regions, if before called setNumThreads with threads > 0, otherwise returns the number of logical CPUs, available for the process.

@sa setNumThreads, getThreadNum

Python prototype (for reference only):

getNumThreads() -> retval
Link to this function

getOptimalDFTSize(vecsize)

View Source
@spec getOptimalDFTSize(integer()) :: integer() | {:error, String.t()}

Returns the optimal DFT size for a given vector size.

Positional Arguments
  • vecsize: int.

    vector size.

Return
  • retval: int

DFT performance is not a monotonic function of a vector size. Therefore, when you calculate convolution of two arrays or perform the spectral analysis of an array, it usually makes sense to pad the input data with zeros to get a bit larger array that can be transformed much faster than the original one. Arrays whose size is a power-of-two (2, 4, 8, 16, 32, ...) are the fastest to process. Though, the arrays whose size is a product of 2's, 3's, and 5's (for example, 300 = 5*5*3*2*2) are also processed quite efficiently. The function cv::getOptimalDFTSize returns the minimum number N that is greater than or equal to vecsize so that the DFT of a vector of size N can be processed efficiently. In the current implementation N = 2 ^p^ * 3 ^q^ * 5 ^r^ for some integer p, q, r. The function returns a negative number if vecsize is too large (very close to INT_MAX ). While the function cannot be used directly to estimate the optimal vector size for DCT transform (since the current DCT implementation supports only even-size vectors), it can be easily processed as getOptimalDFTSize((vecsize+1)/2)*2. @sa dft , dct , idft , idct , mulSpectrums

Python prototype (for reference only):

getOptimalDFTSize(vecsize) -> retval
Link to this function

getOptimalNewCameraMatrix(cameraMatrix, distCoeffs, imageSize, alpha)

View Source
@spec getOptimalNewCameraMatrix(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  number()
) ::
  {Evision.Mat.t(), {number(), number(), number(), number()}}
  | {:error, String.t()}

Returns the new camera intrinsic matrix based on the free scaling parameter.

Positional Arguments
  • cameraMatrix: Evision.Mat.t().

    Input camera intrinsic matrix.

  • distCoeffs: Evision.Mat.t().

    Input vector of distortion coefficients \f$\distcoeffs\f$. If the vector is NULL/empty, the zero distortion coefficients are assumed.

  • imageSize: Size.

    Original image size.

  • alpha: double.

    Free scaling parameter between 0 (when all the pixels in the undistorted image are valid) and 1 (when all the source image pixels are retained in the undistorted image). See #stereoRectify for details.

Keyword Arguments
  • newImgSize: Size.

    Image size after rectification. By default, it is set to imageSize .

  • centerPrincipalPoint: bool.

    Optional flag that indicates whether in the new camera intrinsic matrix the principal point should be at the image center or not. By default, the principal point is chosen to best fit a subset of the source image (determined by alpha) to the corrected image.

Return
  • retval: Evision.Mat.t()

  • validPixROI: Rect*.

    Optional output rectangle that outlines all-good-pixels region in the undistorted image. See roi1, roi2 description in #stereoRectify .

@return new_camera_matrix Output new camera intrinsic matrix. The function computes and returns the optimal new camera intrinsic matrix based on the free scaling parameter. By varying this parameter, you may retrieve only sensible pixels alpha=0 , keep all the original image pixels if there is valuable information in the corners alpha=1 , or get something in between. When alpha>0 , the undistorted result is likely to have some black pixels corresponding to "virtual" pixels outside of the captured distorted image. The original camera intrinsic matrix, distortion coefficients, the computed new camera intrinsic matrix, and newImageSize should be passed to #initUndistortRectifyMap to produce the maps for #remap .

Python prototype (for reference only):

getOptimalNewCameraMatrix(cameraMatrix, distCoeffs, imageSize, alpha[, newImgSize[, centerPrincipalPoint]]) -> retval, validPixROI
Link to this function

getOptimalNewCameraMatrix(cameraMatrix, distCoeffs, imageSize, alpha, opts)

View Source
@spec getOptimalNewCameraMatrix(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  number(),
  [{atom(), term()}, ...] | nil
) ::
  {Evision.Mat.t(), {number(), number(), number(), number()}}
  | {:error, String.t()}

Returns the new camera intrinsic matrix based on the free scaling parameter.

Positional Arguments
  • cameraMatrix: Evision.Mat.t().

    Input camera intrinsic matrix.

  • distCoeffs: Evision.Mat.t().

    Input vector of distortion coefficients \f$\distcoeffs\f$. If the vector is NULL/empty, the zero distortion coefficients are assumed.

  • imageSize: Size.

    Original image size.

  • alpha: double.

    Free scaling parameter between 0 (when all the pixels in the undistorted image are valid) and 1 (when all the source image pixels are retained in the undistorted image). See #stereoRectify for details.

Keyword Arguments
  • newImgSize: Size.

    Image size after rectification. By default, it is set to imageSize .

  • centerPrincipalPoint: bool.

    Optional flag that indicates whether in the new camera intrinsic matrix the principal point should be at the image center or not. By default, the principal point is chosen to best fit a subset of the source image (determined by alpha) to the corrected image.

Return
  • retval: Evision.Mat.t()

  • validPixROI: Rect*.

    Optional output rectangle that outlines all-good-pixels region in the undistorted image. See roi1, roi2 description in #stereoRectify .

@return new_camera_matrix Output new camera intrinsic matrix. The function computes and returns the optimal new camera intrinsic matrix based on the free scaling parameter. By varying this parameter, you may retrieve only sensible pixels alpha=0 , keep all the original image pixels if there is valuable information in the corners alpha=1 , or get something in between. When alpha>0 , the undistorted result is likely to have some black pixels corresponding to "virtual" pixels outside of the captured distorted image. The original camera intrinsic matrix, distortion coefficients, the computed new camera intrinsic matrix, and newImageSize should be passed to #initUndistortRectifyMap to produce the maps for #remap .

Python prototype (for reference only):

getOptimalNewCameraMatrix(cameraMatrix, distCoeffs, imageSize, alpha[, newImgSize[, centerPrincipalPoint]]) -> retval, validPixROI
Link to this function

getPerspectiveTransform(src, dst)

View Source
@spec getPerspectiveTransform(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

Calculates a perspective transform from four pairs of the corresponding points.

Positional Arguments
  • src: Evision.Mat.t().

    Coordinates of quadrangle vertices in the source image.

  • dst: Evision.Mat.t().

    Coordinates of the corresponding quadrangle vertices in the destination image.

Keyword Arguments
  • solveMethod: int.

    method passed to cv::solve (#DecompTypes)

Return
  • retval: Evision.Mat.t()

The function calculates the \f$3 \times 3\f$ matrix of a perspective transform so that: \f[\begin{bmatrix} t_i x'_i \\ t_i y'_i \\ t_i \end{bmatrix} = \texttt{map_matrix} \cdot \begin{bmatrix} x_i \\ y_i \\ 1 \end{bmatrix}\f] where \f[dst(i)=(x'_i,y'_i), src(i)=(x_i, y_i), i=0,1,2,3\f]

@sa findHomography, warpPerspective, perspectiveTransform

Python prototype (for reference only):

getPerspectiveTransform(src, dst[, solveMethod]) -> retval
Link to this function

getPerspectiveTransform(src, dst, opts)

View Source
@spec getPerspectiveTransform(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Calculates a perspective transform from four pairs of the corresponding points.

Positional Arguments
  • src: Evision.Mat.t().

    Coordinates of quadrangle vertices in the source image.

  • dst: Evision.Mat.t().

    Coordinates of the corresponding quadrangle vertices in the destination image.

Keyword Arguments
  • solveMethod: int.

    method passed to cv::solve (#DecompTypes)

Return
  • retval: Evision.Mat.t()

The function calculates the \f$3 \times 3\f$ matrix of a perspective transform so that: \f[\begin{bmatrix} t_i x'_i \\ t_i y'_i \\ t_i \end{bmatrix} = \texttt{map_matrix} \cdot \begin{bmatrix} x_i \\ y_i \\ 1 \end{bmatrix}\f] where \f[dst(i)=(x'_i,y'_i), src(i)=(x_i, y_i), i=0,1,2,3\f]

@sa findHomography, warpPerspective, perspectiveTransform

Python prototype (for reference only):

getPerspectiveTransform(src, dst[, solveMethod]) -> retval
Link to this function

getRectSubPix(image, patchSize, center)

View Source
@spec getRectSubPix(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  {number(), number()}
) ::
  Evision.Mat.t() | {:error, String.t()}

Retrieves a pixel rectangle from an image with sub-pixel accuracy.

Positional Arguments
  • image: Evision.Mat.t().

    Source image.

  • patchSize: Size.

    Size of the extracted patch.

  • center: Point2f.

    Floating point coordinates of the center of the extracted rectangle within the source image. The center must be inside the image.

Keyword Arguments
  • patchType: int.

    Depth of the extracted pixels. By default, they have the same depth as src .

Return
  • patch: Evision.Mat.t().

    Extracted patch that has the size patchSize and the same number of channels as src .

The function getRectSubPix extracts pixels from src: \f[patch(x, y) = src(x + \texttt{center.x} - ( \texttt{dst.cols} -1)*0.5, y + \texttt{center.y} - ( \texttt{dst.rows} -1)*0.5)\f] where the values of the pixels at non-integer coordinates are retrieved using bilinear interpolation. Every channel of multi-channel images is processed independently. Also the image should be a single channel or three channel image. While the center of the rectangle must be inside the image, parts of the rectangle may be outside.

@sa warpAffine, warpPerspective

Python prototype (for reference only):

getRectSubPix(image, patchSize, center[, patch[, patchType]]) -> patch
Link to this function

getRectSubPix(image, patchSize, center, opts)

View Source
@spec getRectSubPix(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  {number(), number()},
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Retrieves a pixel rectangle from an image with sub-pixel accuracy.

Positional Arguments
  • image: Evision.Mat.t().

    Source image.

  • patchSize: Size.

    Size of the extracted patch.

  • center: Point2f.

    Floating point coordinates of the center of the extracted rectangle within the source image. The center must be inside the image.

Keyword Arguments
  • patchType: int.

    Depth of the extracted pixels. By default, they have the same depth as src .

Return
  • patch: Evision.Mat.t().

    Extracted patch that has the size patchSize and the same number of channels as src .

The function getRectSubPix extracts pixels from src: \f[patch(x, y) = src(x + \texttt{center.x} - ( \texttt{dst.cols} -1)*0.5, y + \texttt{center.y} - ( \texttt{dst.rows} -1)*0.5)\f] where the values of the pixels at non-integer coordinates are retrieved using bilinear interpolation. Every channel of multi-channel images is processed independently. Also the image should be a single channel or three channel image. While the center of the rectangle must be inside the image, parts of the rectangle may be outside.

@sa warpAffine, warpPerspective

Python prototype (for reference only):

getRectSubPix(image, patchSize, center[, patch[, patchType]]) -> patch
Link to this function

getRotationMatrix2D(center, angle, scale)

View Source
@spec getRotationMatrix2D({number(), number()}, number(), number()) ::
  Evision.Mat.t() | {:error, String.t()}

Calculates an affine matrix of 2D rotation.

Positional Arguments
  • center: Point2f.

    Center of the rotation in the source image.

  • angle: double.

    Rotation angle in degrees. Positive values mean counter-clockwise rotation (the coordinate origin is assumed to be the top-left corner).

  • scale: double.

    Isotropic scale factor.

Return
  • retval: Evision.Mat.t()

The function calculates the following matrix: \f[\begin{bmatrix} \alpha & \beta & (1- \alpha ) \cdot \texttt{center.x} - \beta \cdot \texttt{center.y} \\ - \beta & \alpha & \beta \cdot \texttt{center.x} + (1- \alpha ) \cdot \texttt{center.y} \end{bmatrix}\f] where \f[\begin{array}{l} \alpha = \texttt{scale} \cdot \cos \texttt{angle} , \\ \beta = \texttt{scale} \cdot \sin \texttt{angle} \end{array}\f] The transformation maps the rotation center to itself. If this is not the target, adjust the shift.

@sa getAffineTransform, warpAffine, transform

Python prototype (for reference only):

getRotationMatrix2D(center, angle, scale) -> retval
Link to this function

getStructuringElement(shape, ksize)

View Source
@spec getStructuringElement(
  integer(),
  {number(), number()}
) :: Evision.Mat.t() | {:error, String.t()}

Returns a structuring element of the specified size and shape for morphological operations.

Positional Arguments
  • shape: int.

    Element shape that could be one of #MorphShapes

  • ksize: Size.

    Size of the structuring element.

Keyword Arguments
  • anchor: Point.

    Anchor position within the element. The default value \f$(-1, -1)\f$ means that the anchor is at the center. Note that only the shape of a cross-shaped element depends on the anchor position. In other cases the anchor just regulates how much the result of the morphological operation is shifted.

Return
  • retval: Evision.Mat.t()

The function constructs and returns the structuring element that can be further passed to #erode, #dilate or #morphologyEx. But you can also construct an arbitrary binary mask yourself and use it as the structuring element.

Python prototype (for reference only):

getStructuringElement(shape, ksize[, anchor]) -> retval
Link to this function

getStructuringElement(shape, ksize, opts)

View Source
@spec getStructuringElement(
  integer(),
  {number(), number()},
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Returns a structuring element of the specified size and shape for morphological operations.

Positional Arguments
  • shape: int.

    Element shape that could be one of #MorphShapes

  • ksize: Size.

    Size of the structuring element.

Keyword Arguments
  • anchor: Point.

    Anchor position within the element. The default value \f$(-1, -1)\f$ means that the anchor is at the center. Note that only the shape of a cross-shaped element depends on the anchor position. In other cases the anchor just regulates how much the result of the morphological operation is shifted.

Return
  • retval: Evision.Mat.t()

The function constructs and returns the structuring element that can be further passed to #erode, #dilate or #morphologyEx. But you can also construct an arbitrary binary mask yourself and use it as the structuring element.

Python prototype (for reference only):

getStructuringElement(shape, ksize[, anchor]) -> retval
Link to this function

getTextSize(text, fontFace, fontScale, thickness)

View Source
@spec getTextSize(binary(), integer(), number(), integer()) ::
  {{number(), number()}, integer()} | {:error, String.t()}

Calculates the width and height of a text string.

Positional Arguments
  • text: String.

    Input text string.

  • fontFace: int.

    Font to use, see #HersheyFonts.

  • fontScale: double.

    Font scale factor that is multiplied by the font-specific base size.

  • thickness: int.

    Thickness of lines used to render the text. See #putText for details.

Return
  • retval: Size

  • baseLine: int*.

    y-coordinate of the baseline relative to the bottom-most text point.

The function cv::getTextSize calculates and returns the size of a box that contains the specified text. That is, the following code renders some text, the tight box surrounding it, and the baseline: :

String text = "Funny text inside the box";
int fontFace = FONT_HERSHEY_SCRIPT_SIMPLEX;
double fontScale = 2;
int thickness = 3;
Mat img(600, 800, CV_8UC3, Scalar::all(0));
int baseline=0;
Size textSize = getTextSize(text, fontFace,
fontScale, thickness, &baseline);
baseline += thickness;
// center the text
Point textOrg((img.cols - textSize.width)/2,
(img.rows + textSize.height)/2);
// draw the box
rectangle(img, textOrg + Point(0, baseline),
textOrg + Point(textSize.width, -textSize.height),
Scalar(0,0,255));
// ... and the baseline first
line(img, textOrg + Point(0, thickness),
textOrg + Point(textSize.width, thickness),
Scalar(0, 0, 255));
// then put the text itself
putText(img, text, textOrg, fontFace, fontScale,
Scalar::all(255), thickness, 8);

@return The size of a box that contains the specified text. @see putText

Python prototype (for reference only):

getTextSize(text, fontFace, fontScale, thickness) -> retval, baseLine
@spec getThreadNum() :: integer() | {:error, String.t()}

Returns the index of the currently executed thread within the current parallel region. Always returns 0 if called outside of parallel region.

Return
  • retval: int

@deprecated Current implementation doesn't corresponding to this documentation. The exact meaning of the return value depends on the threading framework used by OpenCV library:

  • TBB - Unsupported with current 4.1 TBB release. Maybe will be supported in future.

  • OpenMP - The thread number, within the current team, of the calling thread.

  • Concurrency - An ID for the virtual processor that the current context is executing on (0 for master thread and unique number for others, but not necessary 1,2,3,...).

  • GCD - System calling thread's ID. Never returns 0 inside parallel region.

  • C= - The index of the current parallel task. @sa setNumThreads, getNumThreads

Python prototype (for reference only):

getThreadNum() -> retval
@spec getTickCount() :: integer() | {:error, String.t()}

Returns the number of ticks.

Return
  • retval: int64

The function returns the number of ticks after the certain event (for example, when the machine was turned on). It can be used to initialize RNG or to measure a function execution time by reading the tick count before and after the function call. @sa getTickFrequency, TickMeter

Python prototype (for reference only):

getTickCount() -> retval
@spec getTickFrequency() :: number() | {:error, String.t()}

Returns the number of ticks per second.

Return
  • retval: double

The function returns the number of ticks per second. That is, the following code computes the execution time in seconds:

double t = (double)getTickCount();
// do something ...
t = ((double)getTickCount() - t)/getTickFrequency();

@sa getTickCount, TickMeter

Python prototype (for reference only):

getTickFrequency() -> retval
Link to this function

getTrackbarPos(trackbarname, winname)

View Source
@spec getTrackbarPos(binary(), binary()) :: integer() | {:error, String.t()}

Returns the trackbar position.

Positional Arguments
  • trackbarname: String.

    Name of the trackbar.

  • winname: String.

    Name of the window that is the parent of the trackbar.

Return
  • retval: int

The function returns the current position of the specified trackbar. Note: [Qt Backend Only] winname can be empty if the trackbar is attached to the control panel.

Python prototype (for reference only):

getTrackbarPos(trackbarname, winname) -> retval
Link to this function

getValidDisparityROI(roi1, roi2, minDisparity, numberOfDisparities, blockSize)

View Source
@spec getValidDisparityROI(
  {number(), number(), number(), number()},
  {number(), number(), number(), number()},
  integer(),
  integer(),
  integer()
) :: {number(), number(), number(), number()} | {:error, String.t()}

getValidDisparityROI

Positional Arguments
  • roi1: Rect
  • roi2: Rect
  • minDisparity: int
  • numberOfDisparities: int
  • blockSize: int
Return
  • retval: Rect

Python prototype (for reference only):

getValidDisparityROI(roi1, roi2, minDisparity, numberOfDisparities, blockSize) -> retval
@spec getVersionMajor() :: integer() | {:error, String.t()}

Returns major library version

Return
  • retval: int

Python prototype (for reference only):

getVersionMajor() -> retval
@spec getVersionMinor() :: integer() | {:error, String.t()}

Returns minor library version

Return
  • retval: int

Python prototype (for reference only):

getVersionMinor() -> retval
@spec getVersionRevision() :: integer() | {:error, String.t()}

Returns revision field of the library version

Return
  • retval: int

Python prototype (for reference only):

getVersionRevision() -> retval
@spec getVersionString() :: binary() | {:error, String.t()}

Returns library version string

Return

For example "3.4.1-dev". @sa getMajorVersion, getMinorVersion, getRevisionVersion

Python prototype (for reference only):

getVersionString() -> retval
Link to this function

getWindowImageRect(winname)

View Source
@spec getWindowImageRect(binary()) ::
  {number(), number(), number(), number()} | {:error, String.t()}

Provides rectangle of image in the window.

Positional Arguments
  • winname: String.

    Name of the window.

Return
  • retval: Rect

The function getWindowImageRect returns the client screen coordinates, width and height of the image rendering area.

@sa resizeWindow moveWindow

Python prototype (for reference only):

getWindowImageRect(winname) -> retval
Link to this function

getWindowProperty(winname, prop_id)

View Source
@spec getWindowProperty(binary(), integer()) :: number() | {:error, String.t()}

Provides parameters of a window.

Positional Arguments
  • winname: String.

    Name of the window.

  • prop_id: int.

    Window property to retrieve. The following operation flags are available: (cv::WindowPropertyFlags)

Return
  • retval: double

The function getWindowProperty returns properties of a window.

@sa setWindowProperty

Python prototype (for reference only):

getWindowProperty(winname, prop_id) -> retval
Link to this function

goodFeaturesToTrack(image, maxCorners, qualityLevel, minDistance)

View Source
@spec goodFeaturesToTrack(Evision.Mat.maybe_mat_in(), integer(), number(), number()) ::
  Evision.Mat.t() | {:error, String.t()}

Determines strong corners on an image.

Positional Arguments
  • image: Evision.Mat.t().

    Input 8-bit or floating-point 32-bit, single-channel image.

  • maxCorners: int.

    Maximum number of corners to return. If there are more corners than are found, the strongest of them is returned. maxCorners <= 0 implies that no limit on the maximum is set and all detected corners are returned.

  • qualityLevel: double.

    Parameter characterizing the minimal accepted quality of image corners. The parameter value is multiplied by the best corner quality measure, which is the minimal eigenvalue (see #cornerMinEigenVal ) or the Harris function response (see #cornerHarris ). The corners with the quality measure less than the product are rejected. For example, if the best corner has the quality measure = 1500, and the qualityLevel=0.01 , then all the corners with the quality measure less than 15 are rejected.

  • minDistance: double.

    Minimum possible Euclidean distance between the returned corners.

Keyword Arguments
  • mask: Evision.Mat.t().

    Optional region of interest. If the image is not empty (it needs to have the type CV_8UC1 and the same size as image ), it specifies the region in which the corners are detected.

  • blockSize: int.

    Size of an average block for computing a derivative covariation matrix over each pixel neighborhood. See cornerEigenValsAndVecs .

  • useHarrisDetector: bool.

    Parameter indicating whether to use a Harris detector (see #cornerHarris) or #cornerMinEigenVal.

  • k: double.

    Free parameter of the Harris detector.

Return
  • corners: Evision.Mat.t().

    Output vector of detected corners.

The function finds the most prominent corners in the image or in the specified image region, as described in @cite Shi94

  • Function calculates the corner quality measure at every source image pixel using the #cornerMinEigenVal or #cornerHarris .

  • Function performs a non-maximum suppression (the local maximums in 3 x 3 neighborhood are retained).

  • The corners with the minimal eigenvalue less than \f$\texttt{qualityLevel} \cdot \max_{x,y} qualityMeasureMap(x,y)\f$ are rejected.

  • The remaining corners are sorted by the quality measure in the descending order.

  • Function throws away each corner for which there is a stronger corner at a distance less than maxDistance.

The function can be used to initialize a point-based tracker of an object. Note: If the function is called with different values A and B of the parameter qualityLevel , and A > B, the vector of returned corners with qualityLevel=A will be the prefix of the output vector with qualityLevel=B .

@sa cornerMinEigenVal, cornerHarris, calcOpticalFlowPyrLK, estimateRigidTransform,

Python prototype (for reference only):

goodFeaturesToTrack(image, maxCorners, qualityLevel, minDistance[, corners[, mask[, blockSize[, useHarrisDetector[, k]]]]]) -> corners
Link to this function

goodFeaturesToTrack(image, maxCorners, qualityLevel, minDistance, opts)

View Source
@spec goodFeaturesToTrack(
  Evision.Mat.maybe_mat_in(),
  integer(),
  number(),
  number(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Determines strong corners on an image.

Positional Arguments
  • image: Evision.Mat.t().

    Input 8-bit or floating-point 32-bit, single-channel image.

  • maxCorners: int.

    Maximum number of corners to return. If there are more corners than are found, the strongest of them is returned. maxCorners <= 0 implies that no limit on the maximum is set and all detected corners are returned.

  • qualityLevel: double.

    Parameter characterizing the minimal accepted quality of image corners. The parameter value is multiplied by the best corner quality measure, which is the minimal eigenvalue (see #cornerMinEigenVal ) or the Harris function response (see #cornerHarris ). The corners with the quality measure less than the product are rejected. For example, if the best corner has the quality measure = 1500, and the qualityLevel=0.01 , then all the corners with the quality measure less than 15 are rejected.

  • minDistance: double.

    Minimum possible Euclidean distance between the returned corners.

Keyword Arguments
  • mask: Evision.Mat.t().

    Optional region of interest. If the image is not empty (it needs to have the type CV_8UC1 and the same size as image ), it specifies the region in which the corners are detected.

  • blockSize: int.

    Size of an average block for computing a derivative covariation matrix over each pixel neighborhood. See cornerEigenValsAndVecs .

  • useHarrisDetector: bool.

    Parameter indicating whether to use a Harris detector (see #cornerHarris) or #cornerMinEigenVal.

  • k: double.

    Free parameter of the Harris detector.

Return
  • corners: Evision.Mat.t().

    Output vector of detected corners.

The function finds the most prominent corners in the image or in the specified image region, as described in @cite Shi94

  • Function calculates the corner quality measure at every source image pixel using the #cornerMinEigenVal or #cornerHarris .

  • Function performs a non-maximum suppression (the local maximums in 3 x 3 neighborhood are retained).

  • The corners with the minimal eigenvalue less than \f$\texttt{qualityLevel} \cdot \max_{x,y} qualityMeasureMap(x,y)\f$ are rejected.

  • The remaining corners are sorted by the quality measure in the descending order.

  • Function throws away each corner for which there is a stronger corner at a distance less than maxDistance.

The function can be used to initialize a point-based tracker of an object. Note: If the function is called with different values A and B of the parameter qualityLevel , and A > B, the vector of returned corners with qualityLevel=A will be the prefix of the output vector with qualityLevel=B .

@sa cornerMinEigenVal, cornerHarris, calcOpticalFlowPyrLK, estimateRigidTransform,

Python prototype (for reference only):

goodFeaturesToTrack(image, maxCorners, qualityLevel, minDistance[, corners[, mask[, blockSize[, useHarrisDetector[, k]]]]]) -> corners
Link to this function

goodFeaturesToTrack(image, maxCorners, qualityLevel, minDistance, mask, blockSize, gradientSize)

View Source
@spec goodFeaturesToTrack(
  Evision.Mat.maybe_mat_in(),
  integer(),
  number(),
  number(),
  Evision.Mat.maybe_mat_in(),
  integer(),
  integer()
) :: Evision.Mat.t() | {:error, String.t()}

goodFeaturesToTrack

Positional Arguments
  • image: Evision.Mat.t()
  • maxCorners: int
  • qualityLevel: double
  • minDistance: double
  • mask: Evision.Mat.t()
  • blockSize: int
  • gradientSize: int
Keyword Arguments
  • useHarrisDetector: bool.
  • k: double.
Return
  • corners: Evision.Mat.t().

Python prototype (for reference only):

goodFeaturesToTrack(image, maxCorners, qualityLevel, minDistance, mask, blockSize, gradientSize[, corners[, useHarrisDetector[, k]]]) -> corners
Link to this function

goodFeaturesToTrack(image, maxCorners, qualityLevel, minDistance, mask, blockSize, gradientSize, opts)

View Source
@spec goodFeaturesToTrack(
  Evision.Mat.maybe_mat_in(),
  integer(),
  number(),
  number(),
  Evision.Mat.maybe_mat_in(),
  integer(),
  integer(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

goodFeaturesToTrack

Positional Arguments
  • image: Evision.Mat.t()
  • maxCorners: int
  • qualityLevel: double
  • minDistance: double
  • mask: Evision.Mat.t()
  • blockSize: int
  • gradientSize: int
Keyword Arguments
  • useHarrisDetector: bool.
  • k: double.
Return
  • corners: Evision.Mat.t().

Python prototype (for reference only):

goodFeaturesToTrack(image, maxCorners, qualityLevel, minDistance, mask, blockSize, gradientSize[, corners[, useHarrisDetector[, k]]]) -> corners
Link to this function

goodFeaturesToTrackWithQuality(image, maxCorners, qualityLevel, minDistance, mask)

View Source
@spec goodFeaturesToTrackWithQuality(
  Evision.Mat.maybe_mat_in(),
  integer(),
  number(),
  number(),
  Evision.Mat.maybe_mat_in()
) :: {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Same as above, but returns also quality measure of the detected corners.

Positional Arguments
  • image: Evision.Mat.t().

    Input 8-bit or floating-point 32-bit, single-channel image.

  • maxCorners: int.

    Maximum number of corners to return. If there are more corners than are found, the strongest of them is returned. maxCorners <= 0 implies that no limit on the maximum is set and all detected corners are returned.

  • qualityLevel: double.

    Parameter characterizing the minimal accepted quality of image corners. The parameter value is multiplied by the best corner quality measure, which is the minimal eigenvalue (see #cornerMinEigenVal ) or the Harris function response (see #cornerHarris ). The corners with the quality measure less than the product are rejected. For example, if the best corner has the quality measure = 1500, and the qualityLevel=0.01 , then all the corners with the quality measure less than 15 are rejected.

  • minDistance: double.

    Minimum possible Euclidean distance between the returned corners.

  • mask: Evision.Mat.t().

    Region of interest. If the image is not empty (it needs to have the type CV_8UC1 and the same size as image ), it specifies the region in which the corners are detected.

Keyword Arguments
  • blockSize: int.

    Size of an average block for computing a derivative covariation matrix over each pixel neighborhood. See cornerEigenValsAndVecs .

  • gradientSize: int.

    Aperture parameter for the Sobel operator used for derivatives computation. See cornerEigenValsAndVecs .

  • useHarrisDetector: bool.

    Parameter indicating whether to use a Harris detector (see #cornerHarris) or #cornerMinEigenVal.

  • k: double.

    Free parameter of the Harris detector.

Return
  • corners: Evision.Mat.t().

    Output vector of detected corners.

  • cornersQuality: Evision.Mat.t().

    Output vector of quality measure of the detected corners.

Python prototype (for reference only):

goodFeaturesToTrackWithQuality(image, maxCorners, qualityLevel, minDistance, mask[, corners[, cornersQuality[, blockSize[, gradientSize[, useHarrisDetector[, k]]]]]]) -> corners, cornersQuality
Link to this function

goodFeaturesToTrackWithQuality(image, maxCorners, qualityLevel, minDistance, mask, opts)

View Source
@spec goodFeaturesToTrackWithQuality(
  Evision.Mat.maybe_mat_in(),
  integer(),
  number(),
  number(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Same as above, but returns also quality measure of the detected corners.

Positional Arguments
  • image: Evision.Mat.t().

    Input 8-bit or floating-point 32-bit, single-channel image.

  • maxCorners: int.

    Maximum number of corners to return. If there are more corners than are found, the strongest of them is returned. maxCorners <= 0 implies that no limit on the maximum is set and all detected corners are returned.

  • qualityLevel: double.

    Parameter characterizing the minimal accepted quality of image corners. The parameter value is multiplied by the best corner quality measure, which is the minimal eigenvalue (see #cornerMinEigenVal ) or the Harris function response (see #cornerHarris ). The corners with the quality measure less than the product are rejected. For example, if the best corner has the quality measure = 1500, and the qualityLevel=0.01 , then all the corners with the quality measure less than 15 are rejected.

  • minDistance: double.

    Minimum possible Euclidean distance between the returned corners.

  • mask: Evision.Mat.t().

    Region of interest. If the image is not empty (it needs to have the type CV_8UC1 and the same size as image ), it specifies the region in which the corners are detected.

Keyword Arguments
  • blockSize: int.

    Size of an average block for computing a derivative covariation matrix over each pixel neighborhood. See cornerEigenValsAndVecs .

  • gradientSize: int.

    Aperture parameter for the Sobel operator used for derivatives computation. See cornerEigenValsAndVecs .

  • useHarrisDetector: bool.

    Parameter indicating whether to use a Harris detector (see #cornerHarris) or #cornerMinEigenVal.

  • k: double.

    Free parameter of the Harris detector.

Return
  • corners: Evision.Mat.t().

    Output vector of detected corners.

  • cornersQuality: Evision.Mat.t().

    Output vector of quality measure of the detected corners.

Python prototype (for reference only):

goodFeaturesToTrackWithQuality(image, maxCorners, qualityLevel, minDistance, mask[, corners[, cornersQuality[, blockSize[, gradientSize[, useHarrisDetector[, k]]]]]]) -> corners, cornersQuality
Link to this function

grabCut(img, mask, rect, bgdModel, fgdModel, iterCount)

View Source

Runs the GrabCut algorithm.

Positional Arguments
  • img: Evision.Mat.t().

    Input 8-bit 3-channel image.

  • rect: Rect.

    ROI containing a segmented object. The pixels outside of the ROI are marked as "obvious background". The parameter is only used when mode==#GC_INIT_WITH_RECT .

  • iterCount: int.

    Number of iterations the algorithm should make before returning the result. Note that the result can be refined with further calls with mode==#GC_INIT_WITH_MASK or mode==GC_EVAL .

Keyword Arguments
  • mode: int.

    Operation mode that could be one of the #GrabCutModes

Return
  • mask: Evision.Mat.t().

    Input/output 8-bit single-channel mask. The mask is initialized by the function when mode is set to #GC_INIT_WITH_RECT. Its elements may have one of the #GrabCutClasses.

  • bgdModel: Evision.Mat.t().

    Temporary array for the background model. Do not modify it while you are processing the same image.

  • fgdModel: Evision.Mat.t().

    Temporary arrays for the foreground model. Do not modify it while you are processing the same image.

The function implements the GrabCut image segmentation algorithm.

Python prototype (for reference only):

grabCut(img, mask, rect, bgdModel, fgdModel, iterCount[, mode]) -> mask, bgdModel, fgdModel
Link to this function

grabCut(img, mask, rect, bgdModel, fgdModel, iterCount, opts)

View Source

Runs the GrabCut algorithm.

Positional Arguments
  • img: Evision.Mat.t().

    Input 8-bit 3-channel image.

  • rect: Rect.

    ROI containing a segmented object. The pixels outside of the ROI are marked as "obvious background". The parameter is only used when mode==#GC_INIT_WITH_RECT .

  • iterCount: int.

    Number of iterations the algorithm should make before returning the result. Note that the result can be refined with further calls with mode==#GC_INIT_WITH_MASK or mode==GC_EVAL .

Keyword Arguments
  • mode: int.

    Operation mode that could be one of the #GrabCutModes

Return
  • mask: Evision.Mat.t().

    Input/output 8-bit single-channel mask. The mask is initialized by the function when mode is set to #GC_INIT_WITH_RECT. Its elements may have one of the #GrabCutClasses.

  • bgdModel: Evision.Mat.t().

    Temporary array for the background model. Do not modify it while you are processing the same image.

  • fgdModel: Evision.Mat.t().

    Temporary arrays for the foreground model. Do not modify it while you are processing the same image.

The function implements the GrabCut image segmentation algorithm.

Python prototype (for reference only):

grabCut(img, mask, rect, bgdModel, fgdModel, iterCount[, mode]) -> mask, bgdModel, fgdModel
Link to this function

groupRectangles(rectList, groupThreshold)

View Source
@spec groupRectangles([{number(), number(), number(), number()}], integer()) ::
  {[{number(), number(), number(), number()}], [integer()]}
  | {:error, String.t()}

groupRectangles

Positional Arguments
  • groupThreshold: int
Keyword Arguments
  • eps: double.
Return
  • rectList: [Rect]
  • weights: [int]

Has overloading in C++

Python prototype (for reference only):

groupRectangles(rectList, groupThreshold[, eps]) -> rectList, weights
Link to this function

groupRectangles(rectList, groupThreshold, opts)

View Source
@spec groupRectangles(
  [{number(), number(), number(), number()}],
  integer(),
  [{atom(), term()}, ...] | nil
) ::
  {[{number(), number(), number(), number()}], [integer()]}
  | {:error, String.t()}

groupRectangles

Positional Arguments
  • groupThreshold: int
Keyword Arguments
  • eps: double.
Return
  • rectList: [Rect]
  • weights: [int]

Has overloading in C++

Python prototype (for reference only):

groupRectangles(rectList, groupThreshold[, eps]) -> rectList, weights
@spec hasNonZero(Evision.Mat.maybe_mat_in()) :: boolean() | {:error, String.t()}

Checks for the presence of at least one non-zero array element.

Positional Arguments
  • src: Evision.Mat.t().

    single-channel array.

Return
  • retval: bool

The function returns whether there are non-zero elements in src @sa mean, meanStdDev, norm, minMaxLoc, calcCovarMatrix

Python prototype (for reference only):

hasNonZero(src) -> retval
Link to this function

haveImageReader(filename)

View Source
@spec haveImageReader(binary()) :: boolean() | {:error, String.t()}

Returns true if the specified image can be decoded by OpenCV

Positional Arguments
  • filename: String.

    File name of the image

Return
  • retval: bool

Python prototype (for reference only):

haveImageReader(filename) -> retval
Link to this function

haveImageWriter(filename)

View Source
@spec haveImageWriter(binary()) :: boolean() | {:error, String.t()}

Returns true if an image with the specified filename can be encoded by OpenCV

Positional Arguments
  • filename: String.

    File name of the image

Return
  • retval: bool

Python prototype (for reference only):

haveImageWriter(filename) -> retval
@spec haveOpenVX() :: boolean() | {:error, String.t()}

haveOpenVX

Return
  • retval: bool

Python prototype (for reference only):

haveOpenVX() -> retval
@spec hconcat([Evision.Mat.maybe_mat_in()]) :: Evision.Mat.t() | {:error, String.t()}

hconcat

Positional Arguments
  • src: [Evision.Mat].

    input array or vector of matrices. all of the matrices must have the same number of rows and the same depth.

Return
  • dst: Evision.Mat.t().

    output array. It has the same number of rows and depth as the src, and the sum of cols of the src. same depth.

Has overloading in C++

std::vector<cv::Mat> matrices = { cv::Mat(4, 1, CV_8UC1, cv::Scalar(1)),
cv::Mat(4, 1, CV_8UC1, cv::Scalar(2)),
cv::Mat(4, 1, CV_8UC1, cv::Scalar(3)),};
cv::Mat out;
cv::hconcat( matrices, out );
//out:
//[1, 2, 3;
// 1, 2, 3;
// 1, 2, 3;
// 1, 2, 3]

Python prototype (for reference only):

hconcat(src[, dst]) -> dst
@spec hconcat([Evision.Mat.maybe_mat_in()], [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

hconcat

Positional Arguments
  • src: [Evision.Mat].

    input array or vector of matrices. all of the matrices must have the same number of rows and the same depth.

Return
  • dst: Evision.Mat.t().

    output array. It has the same number of rows and depth as the src, and the sum of cols of the src. same depth.

Has overloading in C++

std::vector<cv::Mat> matrices = { cv::Mat(4, 1, CV_8UC1, cv::Scalar(1)),
cv::Mat(4, 1, CV_8UC1, cv::Scalar(2)),
cv::Mat(4, 1, CV_8UC1, cv::Scalar(3)),};
cv::Mat out;
cv::hconcat( matrices, out );
//out:
//[1, 2, 3;
// 1, 2, 3;
// 1, 2, 3;
// 1, 2, 3]

Python prototype (for reference only):

hconcat(src[, dst]) -> dst
Link to this function

houghCircles(image, method, dp, minDist)

View Source
@spec houghCircles(Evision.Mat.maybe_mat_in(), integer(), number(), number()) ::
  Evision.Mat.t() | {:error, String.t()}

Finds circles in a grayscale image using the Hough transform.

Positional Arguments
  • image: Evision.Mat.t().

    8-bit, single-channel, grayscale input image.

  • method: int.

    Detection method, see #HoughModes. The available methods are #HOUGH_GRADIENT and #HOUGH_GRADIENT_ALT.

  • dp: double.

    Inverse ratio of the accumulator resolution to the image resolution. For example, if dp=1 , the accumulator has the same resolution as the input image. If dp=2 , the accumulator has half as big width and height. For #HOUGH_GRADIENT_ALT the recommended value is dp=1.5, unless some small very circles need to be detected.

  • minDist: double.

    Minimum distance between the centers of the detected circles. If the parameter is too small, multiple neighbor circles may be falsely detected in addition to a true one. If it is too large, some circles may be missed.

Keyword Arguments
  • param1: double.

    First method-specific parameter. In case of #HOUGH_GRADIENT and #HOUGH_GRADIENT_ALT, it is the higher threshold of the two passed to the Canny edge detector (the lower one is twice smaller). Note that #HOUGH_GRADIENT_ALT uses #Scharr algorithm to compute image derivatives, so the threshold value should normally be higher, such as 300 or normally exposed and contrasty images.

  • param2: double.

    Second method-specific parameter. In case of #HOUGH_GRADIENT, it is the accumulator threshold for the circle centers at the detection stage. The smaller it is, the more false circles may be detected. Circles, corresponding to the larger accumulator values, will be returned first. In the case of #HOUGH_GRADIENT_ALT algorithm, this is the circle "perfectness" measure. The closer it to 1, the better shaped circles algorithm selects. In most cases 0.9 should be fine. If you want get better detection of small circles, you may decrease it to 0.85, 0.8 or even less. But then also try to limit the search range [minRadius, maxRadius] to avoid many false circles.

  • minRadius: int.

    Minimum circle radius.

  • maxRadius: int.

    Maximum circle radius. If <= 0, uses the maximum image dimension. If < 0, #HOUGH_GRADIENT returns centers without finding the radius. #HOUGH_GRADIENT_ALT always computes circle radiuses.

Return
  • circles: Evision.Mat.t().

    Output vector of found circles. Each vector is encoded as 3 or 4 element floating-point vector \f$(x, y, radius)\f$ or \f$(x, y, radius, votes)\f$ .

The function finds circles in a grayscale image using a modification of the Hough transform. Example: : @include snippets/imgproc_HoughLinesCircles.cpp Note: Usually the function detects the centers of circles well. However, it may fail to find correct radii. You can assist to the function by specifying the radius range ( minRadius and maxRadius ) if you know it. Or, in the case of #HOUGH_GRADIENT method you may set maxRadius to a negative number to return centers only without radius search, and find the correct radius using an additional procedure. It also helps to smooth image a bit unless it's already soft. For example, GaussianBlur() with 7x7 kernel and 1.5x1.5 sigma or similar blurring may help.

@sa fitEllipse, minEnclosingCircle

Python prototype (for reference only):

HoughCircles(image, method, dp, minDist[, circles[, param1[, param2[, minRadius[, maxRadius]]]]]) -> circles
Link to this function

houghCircles(image, method, dp, minDist, opts)

View Source
@spec houghCircles(
  Evision.Mat.maybe_mat_in(),
  integer(),
  number(),
  number(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Finds circles in a grayscale image using the Hough transform.

Positional Arguments
  • image: Evision.Mat.t().

    8-bit, single-channel, grayscale input image.

  • method: int.

    Detection method, see #HoughModes. The available methods are #HOUGH_GRADIENT and #HOUGH_GRADIENT_ALT.

  • dp: double.

    Inverse ratio of the accumulator resolution to the image resolution. For example, if dp=1 , the accumulator has the same resolution as the input image. If dp=2 , the accumulator has half as big width and height. For #HOUGH_GRADIENT_ALT the recommended value is dp=1.5, unless some small very circles need to be detected.

  • minDist: double.

    Minimum distance between the centers of the detected circles. If the parameter is too small, multiple neighbor circles may be falsely detected in addition to a true one. If it is too large, some circles may be missed.

Keyword Arguments
  • param1: double.

    First method-specific parameter. In case of #HOUGH_GRADIENT and #HOUGH_GRADIENT_ALT, it is the higher threshold of the two passed to the Canny edge detector (the lower one is twice smaller). Note that #HOUGH_GRADIENT_ALT uses #Scharr algorithm to compute image derivatives, so the threshold value should normally be higher, such as 300 or normally exposed and contrasty images.

  • param2: double.

    Second method-specific parameter. In case of #HOUGH_GRADIENT, it is the accumulator threshold for the circle centers at the detection stage. The smaller it is, the more false circles may be detected. Circles, corresponding to the larger accumulator values, will be returned first. In the case of #HOUGH_GRADIENT_ALT algorithm, this is the circle "perfectness" measure. The closer it to 1, the better shaped circles algorithm selects. In most cases 0.9 should be fine. If you want get better detection of small circles, you may decrease it to 0.85, 0.8 or even less. But then also try to limit the search range [minRadius, maxRadius] to avoid many false circles.

  • minRadius: int.

    Minimum circle radius.

  • maxRadius: int.

    Maximum circle radius. If <= 0, uses the maximum image dimension. If < 0, #HOUGH_GRADIENT returns centers without finding the radius. #HOUGH_GRADIENT_ALT always computes circle radiuses.

Return
  • circles: Evision.Mat.t().

    Output vector of found circles. Each vector is encoded as 3 or 4 element floating-point vector \f$(x, y, radius)\f$ or \f$(x, y, radius, votes)\f$ .

The function finds circles in a grayscale image using a modification of the Hough transform. Example: : @include snippets/imgproc_HoughLinesCircles.cpp Note: Usually the function detects the centers of circles well. However, it may fail to find correct radii. You can assist to the function by specifying the radius range ( minRadius and maxRadius ) if you know it. Or, in the case of #HOUGH_GRADIENT method you may set maxRadius to a negative number to return centers only without radius search, and find the correct radius using an additional procedure. It also helps to smooth image a bit unless it's already soft. For example, GaussianBlur() with 7x7 kernel and 1.5x1.5 sigma or similar blurring may help.

@sa fitEllipse, minEnclosingCircle

Python prototype (for reference only):

HoughCircles(image, method, dp, minDist[, circles[, param1[, param2[, minRadius[, maxRadius]]]]]) -> circles
Link to this function

houghLines(image, rho, theta, threshold)

View Source
@spec houghLines(Evision.Mat.maybe_mat_in(), number(), number(), integer()) ::
  Evision.Mat.t() | {:error, String.t()}

Finds lines in a binary image using the standard Hough transform.

Positional Arguments
  • image: Evision.Mat.t().

    8-bit, single-channel binary source image. The image may be modified by the function.

  • rho: double.

    Distance resolution of the accumulator in pixels.

  • theta: double.

    Angle resolution of the accumulator in radians.

  • threshold: int.

    %Accumulator threshold parameter. Only those lines are returned that get enough votes ( \f$>\texttt{threshold}\f$ ).

Keyword Arguments
  • srn: double.

    For the multi-scale Hough transform, it is a divisor for the distance resolution rho. The coarse accumulator distance resolution is rho and the accurate accumulator resolution is rho/srn. If both srn=0 and stn=0, the classical Hough transform is used. Otherwise, both these parameters should be positive.

  • stn: double.

    For the multi-scale Hough transform, it is a divisor for the distance resolution theta.

  • min_theta: double.

    For standard and multi-scale Hough transform, minimum angle to check for lines. Must fall between 0 and max_theta.

  • max_theta: double.

    For standard and multi-scale Hough transform, an upper bound for the angle. Must fall between min_theta and CV_PI. The actual maximum angle in the accumulator may be slightly less than max_theta, depending on the parameters min_theta and theta.

Return
  • lines: Evision.Mat.t().

    Output vector of lines. Each line is represented by a 2 or 3 element vector \f$(\rho, \theta)\f$ or \f$(\rho, \theta, \textrm{votes})\f$, where \f$\rho\f$ is the distance from the coordinate origin \f$(0,0)\f$ (top-left corner of the image), \f$\theta\f$ is the line rotation angle in radians ( \f$0 \sim \textrm{vertical line}, \pi/2 \sim \textrm{horizontal line}\f$ ), and \f$\textrm{votes}\f$ is the value of accumulator.

The function implements the standard or standard multi-scale Hough transform algorithm for line detection. See http://homepages.inf.ed.ac.uk/rbf/HIPR2/hough.htm for a good explanation of Hough transform.

Python prototype (for reference only):

HoughLines(image, rho, theta, threshold[, lines[, srn[, stn[, min_theta[, max_theta]]]]]) -> lines
Link to this function

houghLines(image, rho, theta, threshold, opts)

View Source
@spec houghLines(
  Evision.Mat.maybe_mat_in(),
  number(),
  number(),
  integer(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Finds lines in a binary image using the standard Hough transform.

Positional Arguments
  • image: Evision.Mat.t().

    8-bit, single-channel binary source image. The image may be modified by the function.

  • rho: double.

    Distance resolution of the accumulator in pixels.

  • theta: double.

    Angle resolution of the accumulator in radians.

  • threshold: int.

    %Accumulator threshold parameter. Only those lines are returned that get enough votes ( \f$>\texttt{threshold}\f$ ).

Keyword Arguments
  • srn: double.

    For the multi-scale Hough transform, it is a divisor for the distance resolution rho. The coarse accumulator distance resolution is rho and the accurate accumulator resolution is rho/srn. If both srn=0 and stn=0, the classical Hough transform is used. Otherwise, both these parameters should be positive.

  • stn: double.

    For the multi-scale Hough transform, it is a divisor for the distance resolution theta.

  • min_theta: double.

    For standard and multi-scale Hough transform, minimum angle to check for lines. Must fall between 0 and max_theta.

  • max_theta: double.

    For standard and multi-scale Hough transform, an upper bound for the angle. Must fall between min_theta and CV_PI. The actual maximum angle in the accumulator may be slightly less than max_theta, depending on the parameters min_theta and theta.

Return
  • lines: Evision.Mat.t().

    Output vector of lines. Each line is represented by a 2 or 3 element vector \f$(\rho, \theta)\f$ or \f$(\rho, \theta, \textrm{votes})\f$, where \f$\rho\f$ is the distance from the coordinate origin \f$(0,0)\f$ (top-left corner of the image), \f$\theta\f$ is the line rotation angle in radians ( \f$0 \sim \textrm{vertical line}, \pi/2 \sim \textrm{horizontal line}\f$ ), and \f$\textrm{votes}\f$ is the value of accumulator.

The function implements the standard or standard multi-scale Hough transform algorithm for line detection. See http://homepages.inf.ed.ac.uk/rbf/HIPR2/hough.htm for a good explanation of Hough transform.

Python prototype (for reference only):

HoughLines(image, rho, theta, threshold[, lines[, srn[, stn[, min_theta[, max_theta]]]]]) -> lines
Link to this function

houghLinesP(image, rho, theta, threshold)

View Source
@spec houghLinesP(Evision.Mat.maybe_mat_in(), number(), number(), integer()) ::
  Evision.Mat.t() | {:error, String.t()}

Finds line segments in a binary image using the probabilistic Hough transform.

Positional Arguments
  • image: Evision.Mat.t().

    8-bit, single-channel binary source image. The image may be modified by the function.

  • rho: double.

    Distance resolution of the accumulator in pixels.

  • theta: double.

    Angle resolution of the accumulator in radians.

  • threshold: int.

    %Accumulator threshold parameter. Only those lines are returned that get enough votes ( \f$>\texttt{threshold}\f$ ).

Keyword Arguments
  • minLineLength: double.

    Minimum line length. Line segments shorter than that are rejected.

  • maxLineGap: double.

    Maximum allowed gap between points on the same line to link them.

Return
  • lines: Evision.Mat.t().

    Output vector of lines. Each line is represented by a 4-element vector \f$(x_1, y_1, x_2, y_2)\f$ , where \f$(x_1,y_1)\f$ and \f$(x_2, y_2)\f$ are the ending points of each detected line segment.

The function implements the probabilistic Hough transform algorithm for line detection, described in @cite Matas00 See the line detection example below: @include snippets/imgproc_HoughLinesP.cpp This is a sample picture the function parameters have been tuned for: image And this is the output of the above program in case of the probabilistic Hough transform: image

@sa LineSegmentDetector

Python prototype (for reference only):

HoughLinesP(image, rho, theta, threshold[, lines[, minLineLength[, maxLineGap]]]) -> lines
Link to this function

houghLinesP(image, rho, theta, threshold, opts)

View Source
@spec houghLinesP(
  Evision.Mat.maybe_mat_in(),
  number(),
  number(),
  integer(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Finds line segments in a binary image using the probabilistic Hough transform.

Positional Arguments
  • image: Evision.Mat.t().

    8-bit, single-channel binary source image. The image may be modified by the function.

  • rho: double.

    Distance resolution of the accumulator in pixels.

  • theta: double.

    Angle resolution of the accumulator in radians.

  • threshold: int.

    %Accumulator threshold parameter. Only those lines are returned that get enough votes ( \f$>\texttt{threshold}\f$ ).

Keyword Arguments
  • minLineLength: double.

    Minimum line length. Line segments shorter than that are rejected.

  • maxLineGap: double.

    Maximum allowed gap between points on the same line to link them.

Return
  • lines: Evision.Mat.t().

    Output vector of lines. Each line is represented by a 4-element vector \f$(x_1, y_1, x_2, y_2)\f$ , where \f$(x_1,y_1)\f$ and \f$(x_2, y_2)\f$ are the ending points of each detected line segment.

The function implements the probabilistic Hough transform algorithm for line detection, described in @cite Matas00 See the line detection example below: @include snippets/imgproc_HoughLinesP.cpp This is a sample picture the function parameters have been tuned for: image And this is the output of the above program in case of the probabilistic Hough transform: image

@sa LineSegmentDetector

Python prototype (for reference only):

HoughLinesP(image, rho, theta, threshold[, lines[, minLineLength[, maxLineGap]]]) -> lines
Link to this function

houghLinesPointSet(point, lines_max, threshold, min_rho, max_rho, rho_step, min_theta, max_theta, theta_step)

View Source
@spec houghLinesPointSet(
  Evision.Mat.maybe_mat_in(),
  integer(),
  integer(),
  number(),
  number(),
  number(),
  number(),
  number(),
  number()
) :: Evision.Mat.t() | {:error, String.t()}

Finds lines in a set of points using the standard Hough transform.

Positional Arguments
  • point: Evision.Mat.t().

    Input vector of points. Each vector must be encoded as a Point vector \f$(x,y)\f$. Type must be CV_32FC2 or CV_32SC2.

  • lines_max: int.

    Max count of Hough lines.

  • threshold: int.

    %Accumulator threshold parameter. Only those lines are returned that get enough votes ( \f$>\texttt{threshold}\f$ ).

  • min_rho: double.

    Minimum value for \f$\rho\f$ for the accumulator (Note: \f$\rho\f$ can be negative. The absolute value \f$|\rho|\f$ is the distance of a line to the origin.).

  • max_rho: double.

    Maximum value for \f$\rho\f$ for the accumulator.

  • rho_step: double.

    Distance resolution of the accumulator.

  • min_theta: double.

    Minimum angle value of the accumulator in radians.

  • max_theta: double.

    Upper bound for the angle value of the accumulator in radians. The actual maximum angle may be slightly less than max_theta, depending on the parameters min_theta and theta_step.

  • theta_step: double.

    Angle resolution of the accumulator in radians.

Return
  • lines: Evision.Mat.t().

    Output vector of found lines. Each vector is encoded as a vector<Vec3d> \f$(votes, rho, theta)\f$. The larger the value of 'votes', the higher the reliability of the Hough line.

The function finds lines in a set of points using a modification of the Hough transform. @include snippets/imgproc_HoughLinesPointSet.cpp

Python prototype (for reference only):

HoughLinesPointSet(point, lines_max, threshold, min_rho, max_rho, rho_step, min_theta, max_theta, theta_step[, lines]) -> lines
Link to this function

houghLinesPointSet(point, lines_max, threshold, min_rho, max_rho, rho_step, min_theta, max_theta, theta_step, opts)

View Source
@spec houghLinesPointSet(
  Evision.Mat.maybe_mat_in(),
  integer(),
  integer(),
  number(),
  number(),
  number(),
  number(),
  number(),
  number(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Finds lines in a set of points using the standard Hough transform.

Positional Arguments
  • point: Evision.Mat.t().

    Input vector of points. Each vector must be encoded as a Point vector \f$(x,y)\f$. Type must be CV_32FC2 or CV_32SC2.

  • lines_max: int.

    Max count of Hough lines.

  • threshold: int.

    %Accumulator threshold parameter. Only those lines are returned that get enough votes ( \f$>\texttt{threshold}\f$ ).

  • min_rho: double.

    Minimum value for \f$\rho\f$ for the accumulator (Note: \f$\rho\f$ can be negative. The absolute value \f$|\rho|\f$ is the distance of a line to the origin.).

  • max_rho: double.

    Maximum value for \f$\rho\f$ for the accumulator.

  • rho_step: double.

    Distance resolution of the accumulator.

  • min_theta: double.

    Minimum angle value of the accumulator in radians.

  • max_theta: double.

    Upper bound for the angle value of the accumulator in radians. The actual maximum angle may be slightly less than max_theta, depending on the parameters min_theta and theta_step.

  • theta_step: double.

    Angle resolution of the accumulator in radians.

Return
  • lines: Evision.Mat.t().

    Output vector of found lines. Each vector is encoded as a vector<Vec3d> \f$(votes, rho, theta)\f$. The larger the value of 'votes', the higher the reliability of the Hough line.

The function finds lines in a set of points using a modification of the Hough transform. @include snippets/imgproc_HoughLinesPointSet.cpp

Python prototype (for reference only):

HoughLinesPointSet(point, lines_max, threshold, min_rho, max_rho, rho_step, min_theta, max_theta, theta_step[, lines]) -> lines
Link to this function

houghLinesWithAccumulator(image, rho, theta, threshold)

View Source
@spec houghLinesWithAccumulator(
  Evision.Mat.maybe_mat_in(),
  number(),
  number(),
  integer()
) ::
  Evision.Mat.t() | {:error, String.t()}

Finds lines in a binary image using the standard Hough transform and get accumulator.

Positional Arguments
  • image: Evision.Mat.t()
  • rho: double
  • theta: double
  • threshold: int
Keyword Arguments
  • srn: double.
  • stn: double.
  • min_theta: double.
  • max_theta: double.
Return
  • lines: Evision.Mat.t().

Note: This function is for bindings use only. Use original function in C++ code @sa HoughLines

Python prototype (for reference only):

HoughLinesWithAccumulator(image, rho, theta, threshold[, lines[, srn[, stn[, min_theta[, max_theta]]]]]) -> lines
Link to this function

houghLinesWithAccumulator(image, rho, theta, threshold, opts)

View Source
@spec houghLinesWithAccumulator(
  Evision.Mat.maybe_mat_in(),
  number(),
  number(),
  integer(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Finds lines in a binary image using the standard Hough transform and get accumulator.

Positional Arguments
  • image: Evision.Mat.t()
  • rho: double
  • theta: double
  • threshold: int
Keyword Arguments
  • srn: double.
  • stn: double.
  • min_theta: double.
  • max_theta: double.
Return
  • lines: Evision.Mat.t().

Note: This function is for bindings use only. Use original function in C++ code @sa HoughLines

Python prototype (for reference only):

HoughLinesWithAccumulator(image, rho, theta, threshold[, lines[, srn[, stn[, min_theta[, max_theta]]]]]) -> lines
@spec huMoments(map()) :: Evision.Mat.t() | {:error, String.t()}

HuMoments

Positional Arguments
  • m: Moments
Return
  • hu: Evision.Mat.t().

Has overloading in C++

Python prototype (for reference only):

HuMoments(m[, hu]) -> hu
@spec huMoments(map(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

HuMoments

Positional Arguments
  • m: Moments
Return
  • hu: Evision.Mat.t().

Has overloading in C++

Python prototype (for reference only):

HuMoments(m[, hu]) -> hu
@spec idct(Evision.Mat.maybe_mat_in()) :: Evision.Mat.t() | {:error, String.t()}

Calculates the inverse Discrete Cosine Transform of a 1D or 2D array.

Positional Arguments
  • src: Evision.Mat.t().

    input floating-point single-channel array.

Keyword Arguments
  • flags: int.

    operation flags.

Return
  • dst: Evision.Mat.t().

    output array of the same size and type as src.

idct(src, dst, flags) is equivalent to dct(src, dst, flags | DCT_INVERSE). @sa dct, dft, idft, getOptimalDFTSize

Python prototype (for reference only):

idct(src[, dst[, flags]]) -> dst
@spec idct(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Calculates the inverse Discrete Cosine Transform of a 1D or 2D array.

Positional Arguments
  • src: Evision.Mat.t().

    input floating-point single-channel array.

Keyword Arguments
  • flags: int.

    operation flags.

Return
  • dst: Evision.Mat.t().

    output array of the same size and type as src.

idct(src, dst, flags) is equivalent to dct(src, dst, flags | DCT_INVERSE). @sa dct, dft, idft, getOptimalDFTSize

Python prototype (for reference only):

idct(src[, dst[, flags]]) -> dst
@spec idft(Evision.Mat.maybe_mat_in()) :: Evision.Mat.t() | {:error, String.t()}

Calculates the inverse Discrete Fourier Transform of a 1D or 2D array.

Positional Arguments
  • src: Evision.Mat.t().

    input floating-point real or complex array.

Keyword Arguments
  • flags: int.

    operation flags (see dft and #DftFlags).

  • nonzeroRows: int.

    number of dst rows to process; the rest of the rows have undefined content (see the convolution sample in dft description.

Return
  • dst: Evision.Mat.t().

    output array whose size and type depend on the flags.

idft(src, dst, flags) is equivalent to dft(src, dst, flags | #DFT_INVERSE) . Note: None of dft and idft scales the result by default. So, you should pass #DFT_SCALE to one of dft or idft explicitly to make these transforms mutually inverse. @sa dft, dct, idct, mulSpectrums, getOptimalDFTSize

Python prototype (for reference only):

idft(src[, dst[, flags[, nonzeroRows]]]) -> dst
@spec idft(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Calculates the inverse Discrete Fourier Transform of a 1D or 2D array.

Positional Arguments
  • src: Evision.Mat.t().

    input floating-point real or complex array.

Keyword Arguments
  • flags: int.

    operation flags (see dft and #DftFlags).

  • nonzeroRows: int.

    number of dst rows to process; the rest of the rows have undefined content (see the convolution sample in dft description.

Return
  • dst: Evision.Mat.t().

    output array whose size and type depend on the flags.

idft(src, dst, flags) is equivalent to dft(src, dst, flags | #DFT_INVERSE) . Note: None of dft and idft scales the result by default. So, you should pass #DFT_SCALE to one of dft or idft explicitly to make these transforms mutually inverse. @sa dft, dct, idct, mulSpectrums, getOptimalDFTSize

Python prototype (for reference only):

idft(src[, dst[, flags[, nonzeroRows]]]) -> dst
Link to this function

illuminationChange(src, mask)

View Source
@spec illuminationChange(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

Applying an appropriate non-linear transformation to the gradient field inside the selection and then integrating back with a Poisson solver, modifies locally the apparent illumination of an image.

Positional Arguments
  • src: Evision.Mat.t().

    Input 8-bit 3-channel image.

  • mask: Evision.Mat.t().

    Input 8-bit 1 or 3-channel image.

Keyword Arguments
  • alpha: float.

    Value ranges between 0-2.

  • beta: float.

    Value ranges between 0-2.

Return
  • dst: Evision.Mat.t().

    Output image with the same size and type as src.

This is useful to highlight under-exposed foreground objects or to reduce specular reflections.

Python prototype (for reference only):

illuminationChange(src, mask[, dst[, alpha[, beta]]]) -> dst
Link to this function

illuminationChange(src, mask, opts)

View Source
@spec illuminationChange(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Applying an appropriate non-linear transformation to the gradient field inside the selection and then integrating back with a Poisson solver, modifies locally the apparent illumination of an image.

Positional Arguments
  • src: Evision.Mat.t().

    Input 8-bit 3-channel image.

  • mask: Evision.Mat.t().

    Input 8-bit 1 or 3-channel image.

Keyword Arguments
  • alpha: float.

    Value ranges between 0-2.

  • beta: float.

    Value ranges between 0-2.

Return
  • dst: Evision.Mat.t().

    Output image with the same size and type as src.

This is useful to highlight under-exposed foreground objects or to reduce specular reflections.

Python prototype (for reference only):

illuminationChange(src, mask[, dst[, alpha[, beta]]]) -> dst
@spec imcount(binary()) :: integer() | {:error, String.t()}

Returns the number of images inside the give file

Positional Arguments
  • filename: String.

    Name of file to be loaded.

Keyword Arguments
  • flags: int.

    Flag that can take values of cv::ImreadModes, default with cv::IMREAD_ANYCOLOR.

Return
  • retval: size_t

The function imcount will return the number of pages in a multi-page image, or 1 for single-page images

Python prototype (for reference only):

imcount(filename[, flags]) -> retval
@spec imcount(binary(), [{atom(), term()}, ...] | nil) ::
  integer() | {:error, String.t()}

Returns the number of images inside the give file

Positional Arguments
  • filename: String.

    Name of file to be loaded.

Keyword Arguments
  • flags: int.

    Flag that can take values of cv::ImreadModes, default with cv::IMREAD_ANYCOLOR.

Return
  • retval: size_t

The function imcount will return the number of pages in a multi-page image, or 1 for single-page images

Python prototype (for reference only):

imcount(filename[, flags]) -> retval
@spec imdecode(binary(), integer()) :: Evision.Mat.maybe_mat_out()
Link to this function

imdecodemulti(buf, flags)

View Source
@spec imdecodemulti(Evision.Mat.maybe_mat_in(), integer()) ::
  [Evision.Mat.t()] | false | {:error, String.t()}

Reads a multi-page image from a buffer in memory.

Positional Arguments
  • buf: Evision.Mat.t().

    Input array or vector of bytes.

  • flags: int.

    The same flags as in cv::imread, see cv::ImreadModes.

Keyword Arguments
  • range: Range.

    A continuous selection of pages.

Return
  • retval: bool

  • mats: [Evision.Mat].

    A vector of Mat objects holding each page, if more than one.

The function imdecodemulti reads a multi-page image from the specified buffer in the memory. If the buffer is too short or contains invalid data, the function returns false. See cv::imreadmulti for the list of supported formats and flags description. Note: In the case of color images, the decoded images will have the channels stored in B G R order.

Python prototype (for reference only):

imdecodemulti(buf, flags[, mats[, range]]) -> retval, mats
Link to this function

imdecodemulti(buf, flags, opts)

View Source
@spec imdecodemulti(
  Evision.Mat.maybe_mat_in(),
  integer(),
  [{atom(), term()}, ...] | nil
) ::
  [Evision.Mat.t()] | false | {:error, String.t()}

Reads a multi-page image from a buffer in memory.

Positional Arguments
  • buf: Evision.Mat.t().

    Input array or vector of bytes.

  • flags: int.

    The same flags as in cv::imread, see cv::ImreadModes.

Keyword Arguments
  • range: Range.

    A continuous selection of pages.

Return
  • retval: bool

  • mats: [Evision.Mat].

    A vector of Mat objects holding each page, if more than one.

The function imdecodemulti reads a multi-page image from the specified buffer in the memory. If the buffer is too short or contains invalid data, the function returns false. See cv::imreadmulti for the list of supported formats and flags description. Note: In the case of color images, the decoded images will have the channels stored in B G R order.

Python prototype (for reference only):

imdecodemulti(buf, flags[, mats[, range]]) -> retval, mats
@spec imencode(binary(), Evision.Mat.maybe_mat_in()) ::
  binary() | false | {:error, String.t()}

Encodes an image into a memory buffer.

Positional Arguments
  • ext: String.

    File extension that defines the output format. Must include a leading period.

  • img: Evision.Mat.t().

    Image to be written.

Keyword Arguments
  • params: [int].

    Format-specific parameters. See cv::imwrite and cv::ImwriteFlags.

Return
  • retval: bool

  • buf: [uchar].

    Output buffer resized to fit the compressed image.

The function imencode compresses the image and stores it in the memory buffer that is resized to fit the result. See cv::imwrite for the list of supported formats and flags description.

Python prototype (for reference only):

imencode(ext, img[, params]) -> retval, buf
Link to this function

imencode(ext, img, opts)

View Source
@spec imencode(binary(), Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  binary() | false | {:error, String.t()}

Encodes an image into a memory buffer.

Positional Arguments
  • ext: String.

    File extension that defines the output format. Must include a leading period.

  • img: Evision.Mat.t().

    Image to be written.

Keyword Arguments
  • params: [int].

    Format-specific parameters. See cv::imwrite and cv::ImwriteFlags.

Return
  • retval: bool

  • buf: [uchar].

    Output buffer resized to fit the compressed image.

The function imencode compresses the image and stores it in the memory buffer that is resized to fit the result. See cv::imwrite for the list of supported formats and flags description.

Python prototype (for reference only):

imencode(ext, img[, params]) -> retval, buf
@spec imread(binary()) :: Evision.Mat.t() | {:error, String.t()}

Loads an image from a file.

Positional Arguments
  • filename: String.

    Name of file to be loaded.

Keyword Arguments
  • flags: int.

    Flag that can take values of cv::ImreadModes

Return
  • retval: Evision.Mat.t()

@anchor imread The function imread loads an image from the specified file and returns it. If the image cannot be read (because of missing file, improper permissions, unsupported or invalid format), the function returns an empty matrix ( Mat::data==NULL ). Currently, the following file formats are supported:

  • Windows bitmaps - *.bmp, *.dib (always supported)
  • JPEG files - *.jpeg, *.jpg, *.jpe (see the Note section)
  • JPEG 2000 files - *.jp2 (see the Note section)
  • Portable Network Graphics - *.png (see the Note section)
  • WebP - *.webp (see the Note section)
  • AVIF - *.avif (see the Note section)
  • Portable image format - *.pbm, *.pgm, *.ppm *.pxm, *.pnm (always supported)
  • PFM files - *.pfm (see the Note section)
  • Sun rasters - *.sr, *.ras (always supported)
  • TIFF files - *.tiff, *.tif (see the Note section)
  • OpenEXR Image files - *.exr (see the Note section)
  • Radiance HDR - *.hdr, *.pic (always supported)
  • Raster and Vector geospatial data supported by GDAL (see the Note section)

Note:

  • The function determines the type of an image by the content, not by the file extension.

  • In the case of color images, the decoded images will have the channels stored in B G R order.

  • When using IMREAD_GRAYSCALE, the codec's internal grayscale conversion will be used, if available. Results may differ to the output of cvtColor()

  • On Microsoft Windows* OS and MacOSX*, the codecs shipped with an OpenCV image (libjpeg, libpng, libtiff, and libjasper) are used by default. So, OpenCV can always read JPEGs, PNGs, and TIFFs. On MacOSX, there is also an option to use native MacOSX image readers. But beware that currently these native image loaders give images with different pixel values because of the color management embedded into MacOSX.

  • On Linux*, BSD flavors and other Unix-like open-source operating systems, OpenCV looks for codecs supplied with an OS image. Install the relevant packages (do not forget the development files, for example, "libjpeg-dev", in Debian* and Ubuntu*) to get the codec support or turn on the OPENCV_BUILD_3RDPARTY_LIBS flag in CMake.

  • In the case you set WITH_GDAL flag to true in CMake and @ref IMREAD_LOAD_GDAL to load the image, then the GDAL driver will be used in order to decode the image, supporting the following formats: Raster, Vector.

  • If EXIF information is embedded in the image file, the EXIF orientation will be taken into account and thus the image will be rotated accordingly except if the flags @ref IMREAD_IGNORE_ORIENTATION or @ref IMREAD_UNCHANGED are passed.

  • Use the IMREAD_UNCHANGED flag to keep the floating point values from PFM image.

  • By default number of pixels must be less than 2^30. Limit can be set using system variable OPENCV_IO_MAX_IMAGE_PIXELS

Python prototype (for reference only):

imread(filename[, flags]) -> retval
@spec imread(binary(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Loads an image from a file.

Positional Arguments
  • filename: String.

    Name of file to be loaded.

Keyword Arguments
  • flags: int.

    Flag that can take values of cv::ImreadModes

Return
  • retval: Evision.Mat.t()

@anchor imread The function imread loads an image from the specified file and returns it. If the image cannot be read (because of missing file, improper permissions, unsupported or invalid format), the function returns an empty matrix ( Mat::data==NULL ). Currently, the following file formats are supported:

  • Windows bitmaps - *.bmp, *.dib (always supported)
  • JPEG files - *.jpeg, *.jpg, *.jpe (see the Note section)
  • JPEG 2000 files - *.jp2 (see the Note section)
  • Portable Network Graphics - *.png (see the Note section)
  • WebP - *.webp (see the Note section)
  • AVIF - *.avif (see the Note section)
  • Portable image format - *.pbm, *.pgm, *.ppm *.pxm, *.pnm (always supported)
  • PFM files - *.pfm (see the Note section)
  • Sun rasters - *.sr, *.ras (always supported)
  • TIFF files - *.tiff, *.tif (see the Note section)
  • OpenEXR Image files - *.exr (see the Note section)
  • Radiance HDR - *.hdr, *.pic (always supported)
  • Raster and Vector geospatial data supported by GDAL (see the Note section)

Note:

  • The function determines the type of an image by the content, not by the file extension.

  • In the case of color images, the decoded images will have the channels stored in B G R order.

  • When using IMREAD_GRAYSCALE, the codec's internal grayscale conversion will be used, if available. Results may differ to the output of cvtColor()

  • On Microsoft Windows* OS and MacOSX*, the codecs shipped with an OpenCV image (libjpeg, libpng, libtiff, and libjasper) are used by default. So, OpenCV can always read JPEGs, PNGs, and TIFFs. On MacOSX, there is also an option to use native MacOSX image readers. But beware that currently these native image loaders give images with different pixel values because of the color management embedded into MacOSX.

  • On Linux*, BSD flavors and other Unix-like open-source operating systems, OpenCV looks for codecs supplied with an OS image. Install the relevant packages (do not forget the development files, for example, "libjpeg-dev", in Debian* and Ubuntu*) to get the codec support or turn on the OPENCV_BUILD_3RDPARTY_LIBS flag in CMake.

  • In the case you set WITH_GDAL flag to true in CMake and @ref IMREAD_LOAD_GDAL to load the image, then the GDAL driver will be used in order to decode the image, supporting the following formats: Raster, Vector.

  • If EXIF information is embedded in the image file, the EXIF orientation will be taken into account and thus the image will be rotated accordingly except if the flags @ref IMREAD_IGNORE_ORIENTATION or @ref IMREAD_UNCHANGED are passed.

  • Use the IMREAD_UNCHANGED flag to keep the floating point values from PFM image.

  • By default number of pixels must be less than 2^30. Limit can be set using system variable OPENCV_IO_MAX_IMAGE_PIXELS

Python prototype (for reference only):

imread(filename[, flags]) -> retval
@spec imreadmulti(binary()) :: [Evision.Mat.t()] | false | {:error, String.t()}

Loads a multi-page image from a file.

Positional Arguments
  • filename: String.

    Name of file to be loaded.

Keyword Arguments
  • flags: int.

    Flag that can take values of cv::ImreadModes, default with cv::IMREAD_ANYCOLOR.

Return
  • retval: bool

  • mats: [Evision.Mat].

    A vector of Mat objects holding each page.

The function imreadmulti loads a multi-page image from the specified file into a vector of Mat objects. @sa cv::imread

Python prototype (for reference only):

imreadmulti(filename[, mats[, flags]]) -> retval, mats
Link to this function

imreadmulti(filename, opts)

View Source
@spec imreadmulti(binary(), [{atom(), term()}, ...] | nil) ::
  [Evision.Mat.t()] | false | {:error, String.t()}

Loads a multi-page image from a file.

Positional Arguments
  • filename: String.

    Name of file to be loaded.

Keyword Arguments
  • flags: int.

    Flag that can take values of cv::ImreadModes, default with cv::IMREAD_ANYCOLOR.

Return
  • retval: bool

  • mats: [Evision.Mat].

    A vector of Mat objects holding each page.

The function imreadmulti loads a multi-page image from the specified file into a vector of Mat objects. @sa cv::imread

Python prototype (for reference only):

imreadmulti(filename[, mats[, flags]]) -> retval, mats
Link to this function

imreadmulti(filename, start, count)

View Source
@spec imreadmulti(binary(), integer(), integer()) ::
  [Evision.Mat.t()] | false | {:error, String.t()}

Loads a of images of a multi-page image from a file.

Positional Arguments
  • filename: String.

    Name of file to be loaded.

  • start: int.

    Start index of the image to load

  • count: int.

    Count number of images to load

Keyword Arguments
  • flags: int.

    Flag that can take values of cv::ImreadModes, default with cv::IMREAD_ANYCOLOR.

Return
  • retval: bool

  • mats: [Evision.Mat].

    A vector of Mat objects holding each page.

The function imreadmulti loads a specified range from a multi-page image from the specified file into a vector of Mat objects. @sa cv::imread

Python prototype (for reference only):

imreadmulti(filename, start, count[, mats[, flags]]) -> retval, mats
Link to this function

imreadmulti(filename, start, count, opts)

View Source
@spec imreadmulti(binary(), integer(), integer(), [{atom(), term()}, ...] | nil) ::
  [Evision.Mat.t()] | false | {:error, String.t()}

Loads a of images of a multi-page image from a file.

Positional Arguments
  • filename: String.

    Name of file to be loaded.

  • start: int.

    Start index of the image to load

  • count: int.

    Count number of images to load

Keyword Arguments
  • flags: int.

    Flag that can take values of cv::ImreadModes, default with cv::IMREAD_ANYCOLOR.

Return
  • retval: bool

  • mats: [Evision.Mat].

    A vector of Mat objects holding each page.

The function imreadmulti loads a specified range from a multi-page image from the specified file into a vector of Mat objects. @sa cv::imread

Python prototype (for reference only):

imreadmulti(filename, start, count[, mats[, flags]]) -> retval, mats
@spec imwrite(binary(), Evision.Mat.maybe_mat_in()) ::
  boolean() | {:error, String.t()}

Saves an image to a specified file.

Positional Arguments
  • filename: String.

    Name of the file.

  • img: Evision.Mat.t().

    (Mat or vector of Mat) Image or Images to be saved.

Keyword Arguments
  • params: [int].

    Format-specific parameters encoded as pairs (paramId_1, paramValue_1, paramId_2, paramValue_2, ... .) see cv::ImwriteFlags

Return
  • retval: bool

The function imwrite saves the image to the specified file. The image format is chosen based on the filename extension (see cv::imread for the list of extensions). In general, only 8-bit unsigned (CV_8U) single-channel or 3-channel (with 'BGR' channel order) images can be saved using this function, with these exceptions:

  • With OpenEXR encoder, only 32-bit float (CV_32F) images can be saved.

  • 8-bit unsigned (CV_8U) images are not supported.

  • With Radiance HDR encoder, non 64-bit float (CV_64F) images can be saved.

  • All images will be converted to 32-bit float (CV_32F).

  • With JPEG 2000 encoder, 8-bit unsigned (CV_8U) and 16-bit unsigned (CV_16U) images can be saved.

  • With PAM encoder, 8-bit unsigned (CV_8U) and 16-bit unsigned (CV_16U) images can be saved.

  • With PNG encoder, 8-bit unsigned (CV_8U) and 16-bit unsigned (CV_16U) images can be saved.

  • PNG images with an alpha channel can be saved using this function. To do this, create 8-bit (or 16-bit) 4-channel image BGRA, where the alpha channel goes last. Fully transparent pixels should have alpha set to 0, fully opaque pixels should have alpha set to 255/65535 (see the code sample below).

  • With PGM/PPM encoder, 8-bit unsigned (CV_8U) and 16-bit unsigned (CV_16U) images can be saved.

  • With TIFF encoder, 8-bit unsigned (CV_8U), 16-bit unsigned (CV_16U), 32-bit float (CV_32F) and 64-bit float (CV_64F) images can be saved.

  • Multiple images (vector of Mat) can be saved in TIFF format (see the code sample below).

  • 32-bit float 3-channel (CV_32FC3) TIFF images will be saved using the LogLuv high dynamic range encoding (4 bytes per pixel)

If the image format is not supported, the image will be converted to 8-bit unsigned (CV_8U) and saved that way. If the format, depth or channel order is different, use Mat::convertTo and cv::cvtColor to convert it before saving. Or, use the universal FileStorage I/O functions to save the image to XML or YAML format. The sample below shows how to create a BGRA image, how to set custom compression parameters and save it to a PNG file. It also demonstrates how to save multiple images in a TIFF file: @include snippets/imgcodecs_imwrite.cpp

Python prototype (for reference only):

imwrite(filename, img[, params]) -> retval
Link to this function

imwrite(filename, img, opts)

View Source
@spec imwrite(binary(), Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  boolean() | {:error, String.t()}

Saves an image to a specified file.

Positional Arguments
  • filename: String.

    Name of the file.

  • img: Evision.Mat.t().

    (Mat or vector of Mat) Image or Images to be saved.

Keyword Arguments
  • params: [int].

    Format-specific parameters encoded as pairs (paramId_1, paramValue_1, paramId_2, paramValue_2, ... .) see cv::ImwriteFlags

Return
  • retval: bool

The function imwrite saves the image to the specified file. The image format is chosen based on the filename extension (see cv::imread for the list of extensions). In general, only 8-bit unsigned (CV_8U) single-channel or 3-channel (with 'BGR' channel order) images can be saved using this function, with these exceptions:

  • With OpenEXR encoder, only 32-bit float (CV_32F) images can be saved.

  • 8-bit unsigned (CV_8U) images are not supported.

  • With Radiance HDR encoder, non 64-bit float (CV_64F) images can be saved.

  • All images will be converted to 32-bit float (CV_32F).

  • With JPEG 2000 encoder, 8-bit unsigned (CV_8U) and 16-bit unsigned (CV_16U) images can be saved.

  • With PAM encoder, 8-bit unsigned (CV_8U) and 16-bit unsigned (CV_16U) images can be saved.

  • With PNG encoder, 8-bit unsigned (CV_8U) and 16-bit unsigned (CV_16U) images can be saved.

  • PNG images with an alpha channel can be saved using this function. To do this, create 8-bit (or 16-bit) 4-channel image BGRA, where the alpha channel goes last. Fully transparent pixels should have alpha set to 0, fully opaque pixels should have alpha set to 255/65535 (see the code sample below).

  • With PGM/PPM encoder, 8-bit unsigned (CV_8U) and 16-bit unsigned (CV_16U) images can be saved.

  • With TIFF encoder, 8-bit unsigned (CV_8U), 16-bit unsigned (CV_16U), 32-bit float (CV_32F) and 64-bit float (CV_64F) images can be saved.

  • Multiple images (vector of Mat) can be saved in TIFF format (see the code sample below).

  • 32-bit float 3-channel (CV_32FC3) TIFF images will be saved using the LogLuv high dynamic range encoding (4 bytes per pixel)

If the image format is not supported, the image will be converted to 8-bit unsigned (CV_8U) and saved that way. If the format, depth or channel order is different, use Mat::convertTo and cv::cvtColor to convert it before saving. Or, use the universal FileStorage I/O functions to save the image to XML or YAML format. The sample below shows how to create a BGRA image, how to set custom compression parameters and save it to a PNG file. It also demonstrates how to save multiple images in a TIFF file: @include snippets/imgcodecs_imwrite.cpp

Python prototype (for reference only):

imwrite(filename, img[, params]) -> retval
Link to this function

imwritemulti(filename, img)

View Source
@spec imwritemulti(binary(), [Evision.Mat.maybe_mat_in()]) ::
  boolean() | {:error, String.t()}

imwritemulti

Positional Arguments
  • filename: String
  • img: [Evision.Mat]
Keyword Arguments
  • params: [int].
Return
  • retval: bool

Python prototype (for reference only):

imwritemulti(filename, img[, params]) -> retval
Link to this function

imwritemulti(filename, img, opts)

View Source
@spec imwritemulti(
  binary(),
  [Evision.Mat.maybe_mat_in()],
  [{atom(), term()}, ...] | nil
) ::
  boolean() | {:error, String.t()}

imwritemulti

Positional Arguments
  • filename: String
  • img: [Evision.Mat]
Keyword Arguments
  • params: [int].
Return
  • retval: bool

Python prototype (for reference only):

imwritemulti(filename, img[, params]) -> retval
Link to this function

initCameraMatrix2D(objectPoints, imagePoints, imageSize)

View Source
@spec initCameraMatrix2D(
  [Evision.Mat.maybe_mat_in()],
  [Evision.Mat.maybe_mat_in()],
  {number(), number()}
) :: Evision.Mat.t() | {:error, String.t()}

Finds an initial camera intrinsic matrix from 3D-2D point correspondences.

Positional Arguments
  • objectPoints: [Evision.Mat].

    Vector of vectors of the calibration pattern points in the calibration pattern coordinate space. In the old interface all the per-view vectors are concatenated. See #calibrateCamera for details.

  • imagePoints: [Evision.Mat].

    Vector of vectors of the projections of the calibration pattern points. In the old interface all the per-view vectors are concatenated.

  • imageSize: Size.

    Image size in pixels used to initialize the principal point.

Keyword Arguments
  • aspectRatio: double.

    If it is zero or negative, both \f$f_x\f$ and \f$f_y\f$ are estimated independently. Otherwise, \f$f_x = f_y \cdot \texttt{aspectRatio}\f$ .

Return
  • retval: Evision.Mat.t()

The function estimates and returns an initial camera intrinsic matrix for the camera calibration process. Currently, the function only supports planar calibration patterns, which are patterns where each object point has z-coordinate =0.

Python prototype (for reference only):

initCameraMatrix2D(objectPoints, imagePoints, imageSize[, aspectRatio]) -> retval
Link to this function

initCameraMatrix2D(objectPoints, imagePoints, imageSize, opts)

View Source
@spec initCameraMatrix2D(
  [Evision.Mat.maybe_mat_in()],
  [Evision.Mat.maybe_mat_in()],
  {number(), number()},
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Finds an initial camera intrinsic matrix from 3D-2D point correspondences.

Positional Arguments
  • objectPoints: [Evision.Mat].

    Vector of vectors of the calibration pattern points in the calibration pattern coordinate space. In the old interface all the per-view vectors are concatenated. See #calibrateCamera for details.

  • imagePoints: [Evision.Mat].

    Vector of vectors of the projections of the calibration pattern points. In the old interface all the per-view vectors are concatenated.

  • imageSize: Size.

    Image size in pixels used to initialize the principal point.

Keyword Arguments
  • aspectRatio: double.

    If it is zero or negative, both \f$f_x\f$ and \f$f_y\f$ are estimated independently. Otherwise, \f$f_x = f_y \cdot \texttt{aspectRatio}\f$ .

Return
  • retval: Evision.Mat.t()

The function estimates and returns an initial camera intrinsic matrix for the camera calibration process. Currently, the function only supports planar calibration patterns, which are patterns where each object point has z-coordinate =0.

Python prototype (for reference only):

initCameraMatrix2D(objectPoints, imagePoints, imageSize[, aspectRatio]) -> retval
Link to this function

initInverseRectificationMap(cameraMatrix, distCoeffs, r, newCameraMatrix, size, m1type)

View Source

Computes the projection and inverse-rectification transformation map. In essense, this is the inverse of #initUndistortRectifyMap to accomodate stereo-rectification of projectors ('inverse-cameras') in projector-camera pairs.

Positional Arguments
  • cameraMatrix: Evision.Mat.t().

    Input camera matrix \f$A=\vecthreethree{f_x}{0}{c_x}{0}{f_y}{c_y}{0}{0}{1}\f$ .

  • distCoeffs: Evision.Mat.t().

    Input vector of distortion coefficients \f$(k_1, k_2, p_1, p_2[, k_3[, k_4, k_5, k_6[, s_1, s_2, s_3, s_4[, \tau_x, \tau_y]]]])\f$ of 4, 5, 8, 12 or 14 elements. If the vector is NULL/empty, the zero distortion coefficients are assumed.

  • r: Evision.Mat.t().

    Optional rectification transformation in the object space (3x3 matrix). R1 or R2, computed by #stereoRectify can be passed here. If the matrix is empty, the identity transformation is assumed.

  • newCameraMatrix: Evision.Mat.t().

    New camera matrix \f$A'=\vecthreethree{f_x'}{0}{c_x'}{0}{f_y'}{c_y'}{0}{0}{1}\f$.

  • size: Size.

    Distorted image size.

  • m1type: int.

    Type of the first output map. Can be CV_32FC1, CV_32FC2 or CV_16SC2, see #convertMaps

Return
  • map1: Evision.Mat.t().

    The first output map for #remap.

  • map2: Evision.Mat.t().

    The second output map for #remap.

The function computes the joint projection and inverse rectification transformation and represents the result in the form of maps for #remap. The projected image looks like a distorted version of the original which, once projected by a projector, should visually match the original. In case of a monocular camera, newCameraMatrix is usually equal to cameraMatrix, or it can be computed by #getOptimalNewCameraMatrix for a better control over scaling. In case of a projector-camera pair, newCameraMatrix is normally set to P1 or P2 computed by #stereoRectify . The projector is oriented differently in the coordinate space, according to R. In case of projector-camera pairs, this helps align the projector (in the same manner as #initUndistortRectifyMap for the camera) to create a stereo-rectified pair. This allows epipolar lines on both images to become horizontal and have the same y-coordinate (in case of a horizontally aligned projector-camera pair). The function builds the maps for the inverse mapping algorithm that is used by #remap. That is, for each pixel \f$(u, v)\f$ in the destination (projected and inverse-rectified) image, the function computes the corresponding coordinates in the source image (that is, in the original digital image). The following process is applied: \f[ \begin{array}{l} \text{newCameraMatrix}\\ x \leftarrow (u - {c'}_x)/{f'}_x \\ y \leftarrow (v - {c'}_y)/{f'}_y \\ \\\text{Undistortion} \\\scriptsize{\textit{though equation shown is for radial undistortion, function implements cv::undistortPoints()}}\\ r^2 \leftarrow x^2 + y^2 \\ \theta \leftarrow \frac{1 + k_1 r^2 + k_2 r^4 + k_3 r^6}{1 + k_4 r^2 + k_5 r^4 + k_6 r^6}\\ x' \leftarrow \frac{x}{\theta} \\ y' \leftarrow \frac{y}{\theta} \\ \\\text{Rectification}\\ {[X\,Y\,W]} ^T \leftarrow R*[x' \, y' \, 1]^T \\ x'' \leftarrow X/W \\ y'' \leftarrow Y/W \\ \\\text{cameraMatrix}\\ map_x(u,v) \leftarrow x'' f_x + c_x \\ map_y(u,v) \leftarrow y'' f_y + c_y \end{array} \f] where \f$(k_1, k_2, p_1, p_2[, k_3[, k_4, k_5, k_6[, s_1, s_2, s_3, s_4[, \tau_x, \tau_y]]]])\f$ are the distortion coefficients vector distCoeffs. In case of a stereo-rectified projector-camera pair, this function is called for the projector while #initUndistortRectifyMap is called for the camera head. This is done after #stereoRectify, which in turn is called after #stereoCalibrate. If the projector-camera pair is not calibrated, it is still possible to compute the rectification transformations directly from the fundamental matrix using #stereoRectifyUncalibrated. For the projector and camera, the function computes homography H as the rectification transformation in a pixel domain, not a rotation matrix R in 3D space. R can be computed from H as \f[\texttt{R} = \texttt{cameraMatrix} ^{-1} \cdot \texttt{H} \cdot \texttt{cameraMatrix}\f] where cameraMatrix can be chosen arbitrarily.

Python prototype (for reference only):

initInverseRectificationMap(cameraMatrix, distCoeffs, R, newCameraMatrix, size, m1type[, map1[, map2]]) -> map1, map2
Link to this function

initInverseRectificationMap(cameraMatrix, distCoeffs, r, newCameraMatrix, size, m1type, opts)

View Source
@spec initInverseRectificationMap(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  integer(),
  [{atom(), term()}, ...] | nil
) :: {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Computes the projection and inverse-rectification transformation map. In essense, this is the inverse of #initUndistortRectifyMap to accomodate stereo-rectification of projectors ('inverse-cameras') in projector-camera pairs.

Positional Arguments
  • cameraMatrix: Evision.Mat.t().

    Input camera matrix \f$A=\vecthreethree{f_x}{0}{c_x}{0}{f_y}{c_y}{0}{0}{1}\f$ .

  • distCoeffs: Evision.Mat.t().

    Input vector of distortion coefficients \f$(k_1, k_2, p_1, p_2[, k_3[, k_4, k_5, k_6[, s_1, s_2, s_3, s_4[, \tau_x, \tau_y]]]])\f$ of 4, 5, 8, 12 or 14 elements. If the vector is NULL/empty, the zero distortion coefficients are assumed.

  • r: Evision.Mat.t().

    Optional rectification transformation in the object space (3x3 matrix). R1 or R2, computed by #stereoRectify can be passed here. If the matrix is empty, the identity transformation is assumed.

  • newCameraMatrix: Evision.Mat.t().

    New camera matrix \f$A'=\vecthreethree{f_x'}{0}{c_x'}{0}{f_y'}{c_y'}{0}{0}{1}\f$.

  • size: Size.

    Distorted image size.

  • m1type: int.

    Type of the first output map. Can be CV_32FC1, CV_32FC2 or CV_16SC2, see #convertMaps

Return
  • map1: Evision.Mat.t().

    The first output map for #remap.

  • map2: Evision.Mat.t().

    The second output map for #remap.

The function computes the joint projection and inverse rectification transformation and represents the result in the form of maps for #remap. The projected image looks like a distorted version of the original which, once projected by a projector, should visually match the original. In case of a monocular camera, newCameraMatrix is usually equal to cameraMatrix, or it can be computed by #getOptimalNewCameraMatrix for a better control over scaling. In case of a projector-camera pair, newCameraMatrix is normally set to P1 or P2 computed by #stereoRectify . The projector is oriented differently in the coordinate space, according to R. In case of projector-camera pairs, this helps align the projector (in the same manner as #initUndistortRectifyMap for the camera) to create a stereo-rectified pair. This allows epipolar lines on both images to become horizontal and have the same y-coordinate (in case of a horizontally aligned projector-camera pair). The function builds the maps for the inverse mapping algorithm that is used by #remap. That is, for each pixel \f$(u, v)\f$ in the destination (projected and inverse-rectified) image, the function computes the corresponding coordinates in the source image (that is, in the original digital image). The following process is applied: \f[ \begin{array}{l} \text{newCameraMatrix}\\ x \leftarrow (u - {c'}_x)/{f'}_x \\ y \leftarrow (v - {c'}_y)/{f'}_y \\ \\\text{Undistortion} \\\scriptsize{\textit{though equation shown is for radial undistortion, function implements cv::undistortPoints()}}\\ r^2 \leftarrow x^2 + y^2 \\ \theta \leftarrow \frac{1 + k_1 r^2 + k_2 r^4 + k_3 r^6}{1 + k_4 r^2 + k_5 r^4 + k_6 r^6}\\ x' \leftarrow \frac{x}{\theta} \\ y' \leftarrow \frac{y}{\theta} \\ \\\text{Rectification}\\ {[X\,Y\,W]} ^T \leftarrow R*[x' \, y' \, 1]^T \\ x'' \leftarrow X/W \\ y'' \leftarrow Y/W \\ \\\text{cameraMatrix}\\ map_x(u,v) \leftarrow x'' f_x + c_x \\ map_y(u,v) \leftarrow y'' f_y + c_y \end{array} \f] where \f$(k_1, k_2, p_1, p_2[, k_3[, k_4, k_5, k_6[, s_1, s_2, s_3, s_4[, \tau_x, \tau_y]]]])\f$ are the distortion coefficients vector distCoeffs. In case of a stereo-rectified projector-camera pair, this function is called for the projector while #initUndistortRectifyMap is called for the camera head. This is done after #stereoRectify, which in turn is called after #stereoCalibrate. If the projector-camera pair is not calibrated, it is still possible to compute the rectification transformations directly from the fundamental matrix using #stereoRectifyUncalibrated. For the projector and camera, the function computes homography H as the rectification transformation in a pixel domain, not a rotation matrix R in 3D space. R can be computed from H as \f[\texttt{R} = \texttt{cameraMatrix} ^{-1} \cdot \texttt{H} \cdot \texttt{cameraMatrix}\f] where cameraMatrix can be chosen arbitrarily.

Python prototype (for reference only):

initInverseRectificationMap(cameraMatrix, distCoeffs, R, newCameraMatrix, size, m1type[, map1[, map2]]) -> map1, map2
Link to this function

initUndistortRectifyMap(cameraMatrix, distCoeffs, r, newCameraMatrix, size, m1type)

View Source

Computes the undistortion and rectification transformation map.

Positional Arguments
  • cameraMatrix: Evision.Mat.t().

    Input camera matrix \f$A=\vecthreethree{f_x}{0}{c_x}{0}{f_y}{c_y}{0}{0}{1}\f$ .

  • distCoeffs: Evision.Mat.t().

    Input vector of distortion coefficients \f$(k_1, k_2, p_1, p_2[, k_3[, k_4, k_5, k_6[, s_1, s_2, s_3, s_4[, \tau_x, \tau_y]]]])\f$ of 4, 5, 8, 12 or 14 elements. If the vector is NULL/empty, the zero distortion coefficients are assumed.

  • r: Evision.Mat.t().

    Optional rectification transformation in the object space (3x3 matrix). R1 or R2 , computed by #stereoRectify can be passed here. If the matrix is empty, the identity transformation is assumed. In #initUndistortRectifyMap R assumed to be an identity matrix.

  • newCameraMatrix: Evision.Mat.t().

    New camera matrix \f$A'=\vecthreethree{f_x'}{0}{c_x'}{0}{f_y'}{c_y'}{0}{0}{1}\f$.

  • size: Size.

    Undistorted image size.

  • m1type: int.

    Type of the first output map that can be CV_32FC1, CV_32FC2 or CV_16SC2, see #convertMaps

Return
  • map1: Evision.Mat.t().

    The first output map.

  • map2: Evision.Mat.t().

    The second output map.

The function computes the joint undistortion and rectification transformation and represents the result in the form of maps for #remap. The undistorted image looks like original, as if it is captured with a camera using the camera matrix =newCameraMatrix and zero distortion. In case of a monocular camera, newCameraMatrix is usually equal to cameraMatrix, or it can be computed by #getOptimalNewCameraMatrix for a better control over scaling. In case of a stereo camera, newCameraMatrix is normally set to P1 or P2 computed by #stereoRectify . Also, this new camera is oriented differently in the coordinate space, according to R. That, for example, helps to align two heads of a stereo camera so that the epipolar lines on both images become horizontal and have the same y- coordinate (in case of a horizontally aligned stereo camera). The function actually builds the maps for the inverse mapping algorithm that is used by #remap. That is, for each pixel \f$(u, v)\f$ in the destination (corrected and rectified) image, the function computes the corresponding coordinates in the source image (that is, in the original image from camera). The following process is applied: \f[ \begin{array}{l} x \leftarrow (u - {c'}_x)/{f'}_x \\ y \leftarrow (v - {c'}_y)/{f'}_y \\ {[X\,Y\,W]} ^T \leftarrow R^{-1}*[x \, y \, 1]^T \\ x' \leftarrow X/W \\ y' \leftarrow Y/W \\ r^2 \leftarrow x'^2 + y'^2 \\ x'' \leftarrow x' \frac{1 + k_1 r^2 + k_2 r^4 + k_3 r^6}{1 + k_4 r^2 + k_5 r^4 + k_6 r^6} + 2p_1 x' y' + p_2(r^2 + 2 x'^2) + s_1 r^2 + s_2 r^4\\ y'' \leftarrow y' \frac{1 + k_1 r^2 + k_2 r^4 + k_3 r^6}{1 + k_4 r^2 + k_5 r^4 + k_6 r^6} + p_1 (r^2 + 2 y'^2) + 2 p_2 x' y' + s_3 r^2 + s_4 r^4 \\ s\vecthree{x'''}{y'''}{1} = \vecthreethree{R_{33}(\tau_x, \tau_y)}{0}{-R_{13}((\tau_x, \tau_y)} {0}{R_{33}(\tau_x, \tau_y)}{-R_{23}(\tau_x, \tau_y)} {0}{0}{1} R(\tau_x, \tau_y) \vecthree{x''}{y''}{1}\\ map_x(u,v) \leftarrow x''' f_x + c_x \\ map_y(u,v) \leftarrow y''' f_y + c_y \end{array} \f] where \f$(k_1, k_2, p_1, p_2[, k_3[, k_4, k_5, k_6[, s_1, s_2, s_3, s_4[, \tau_x, \tau_y]]]])\f$ are the distortion coefficients. In case of a stereo camera, this function is called twice: once for each camera head, after #stereoRectify, which in its turn is called after #stereoCalibrate. But if the stereo camera was not calibrated, it is still possible to compute the rectification transformations directly from the fundamental matrix using #stereoRectifyUncalibrated. For each camera, the function computes homography H as the rectification transformation in a pixel domain, not a rotation matrix R in 3D space. R can be computed from H as \f[\texttt{R} = \texttt{cameraMatrix} ^{-1} \cdot \texttt{H} \cdot \texttt{cameraMatrix}\f] where cameraMatrix can be chosen arbitrarily.

Python prototype (for reference only):

initUndistortRectifyMap(cameraMatrix, distCoeffs, R, newCameraMatrix, size, m1type[, map1[, map2]]) -> map1, map2
Link to this function

initUndistortRectifyMap(cameraMatrix, distCoeffs, r, newCameraMatrix, size, m1type, opts)

View Source
@spec initUndistortRectifyMap(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  integer(),
  [{atom(), term()}, ...] | nil
) :: {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Computes the undistortion and rectification transformation map.

Positional Arguments
  • cameraMatrix: Evision.Mat.t().

    Input camera matrix \f$A=\vecthreethree{f_x}{0}{c_x}{0}{f_y}{c_y}{0}{0}{1}\f$ .

  • distCoeffs: Evision.Mat.t().

    Input vector of distortion coefficients \f$(k_1, k_2, p_1, p_2[, k_3[, k_4, k_5, k_6[, s_1, s_2, s_3, s_4[, \tau_x, \tau_y]]]])\f$ of 4, 5, 8, 12 or 14 elements. If the vector is NULL/empty, the zero distortion coefficients are assumed.

  • r: Evision.Mat.t().

    Optional rectification transformation in the object space (3x3 matrix). R1 or R2 , computed by #stereoRectify can be passed here. If the matrix is empty, the identity transformation is assumed. In #initUndistortRectifyMap R assumed to be an identity matrix.

  • newCameraMatrix: Evision.Mat.t().

    New camera matrix \f$A'=\vecthreethree{f_x'}{0}{c_x'}{0}{f_y'}{c_y'}{0}{0}{1}\f$.

  • size: Size.

    Undistorted image size.

  • m1type: int.

    Type of the first output map that can be CV_32FC1, CV_32FC2 or CV_16SC2, see #convertMaps

Return
  • map1: Evision.Mat.t().

    The first output map.

  • map2: Evision.Mat.t().

    The second output map.

The function computes the joint undistortion and rectification transformation and represents the result in the form of maps for #remap. The undistorted image looks like original, as if it is captured with a camera using the camera matrix =newCameraMatrix and zero distortion. In case of a monocular camera, newCameraMatrix is usually equal to cameraMatrix, or it can be computed by #getOptimalNewCameraMatrix for a better control over scaling. In case of a stereo camera, newCameraMatrix is normally set to P1 or P2 computed by #stereoRectify . Also, this new camera is oriented differently in the coordinate space, according to R. That, for example, helps to align two heads of a stereo camera so that the epipolar lines on both images become horizontal and have the same y- coordinate (in case of a horizontally aligned stereo camera). The function actually builds the maps for the inverse mapping algorithm that is used by #remap. That is, for each pixel \f$(u, v)\f$ in the destination (corrected and rectified) image, the function computes the corresponding coordinates in the source image (that is, in the original image from camera). The following process is applied: \f[ \begin{array}{l} x \leftarrow (u - {c'}_x)/{f'}_x \\ y \leftarrow (v - {c'}_y)/{f'}_y \\ {[X\,Y\,W]} ^T \leftarrow R^{-1}*[x \, y \, 1]^T \\ x' \leftarrow X/W \\ y' \leftarrow Y/W \\ r^2 \leftarrow x'^2 + y'^2 \\ x'' \leftarrow x' \frac{1 + k_1 r^2 + k_2 r^4 + k_3 r^6}{1 + k_4 r^2 + k_5 r^4 + k_6 r^6} + 2p_1 x' y' + p_2(r^2 + 2 x'^2) + s_1 r^2 + s_2 r^4\\ y'' \leftarrow y' \frac{1 + k_1 r^2 + k_2 r^4 + k_3 r^6}{1 + k_4 r^2 + k_5 r^4 + k_6 r^6} + p_1 (r^2 + 2 y'^2) + 2 p_2 x' y' + s_3 r^2 + s_4 r^4 \\ s\vecthree{x'''}{y'''}{1} = \vecthreethree{R_{33}(\tau_x, \tau_y)}{0}{-R_{13}((\tau_x, \tau_y)} {0}{R_{33}(\tau_x, \tau_y)}{-R_{23}(\tau_x, \tau_y)} {0}{0}{1} R(\tau_x, \tau_y) \vecthree{x''}{y''}{1}\\ map_x(u,v) \leftarrow x''' f_x + c_x \\ map_y(u,v) \leftarrow y''' f_y + c_y \end{array} \f] where \f$(k_1, k_2, p_1, p_2[, k_3[, k_4, k_5, k_6[, s_1, s_2, s_3, s_4[, \tau_x, \tau_y]]]])\f$ are the distortion coefficients. In case of a stereo camera, this function is called twice: once for each camera head, after #stereoRectify, which in its turn is called after #stereoCalibrate. But if the stereo camera was not calibrated, it is still possible to compute the rectification transformations directly from the fundamental matrix using #stereoRectifyUncalibrated. For each camera, the function computes homography H as the rectification transformation in a pixel domain, not a rotation matrix R in 3D space. R can be computed from H as \f[\texttt{R} = \texttt{cameraMatrix} ^{-1} \cdot \texttt{H} \cdot \texttt{cameraMatrix}\f] where cameraMatrix can be chosen arbitrarily.

Python prototype (for reference only):

initUndistortRectifyMap(cameraMatrix, distCoeffs, R, newCameraMatrix, size, m1type[, map1[, map2]]) -> map1, map2
Link to this function

inpaint(src, inpaintMask, inpaintRadius, flags)

View Source

Restores the selected region in an image using the region neighborhood.

Positional Arguments
  • src: Evision.Mat.t().

    Input 8-bit, 16-bit unsigned or 32-bit float 1-channel or 8-bit 3-channel image.

  • inpaintMask: Evision.Mat.t().

    Inpainting mask, 8-bit 1-channel image. Non-zero pixels indicate the area that needs to be inpainted.

  • inpaintRadius: double.

    Radius of a circular neighborhood of each point inpainted that is considered by the algorithm.

  • flags: int.

    Inpainting method that could be cv::INPAINT_NS or cv::INPAINT_TELEA

Return
  • dst: Evision.Mat.t().

    Output image with the same size and type as src .

The function reconstructs the selected image area from the pixel near the area boundary. The function may be used to remove dust and scratches from a scanned photo, or to remove undesirable objects from still images or video. See http://en.wikipedia.org/wiki/Inpainting for more details. Note:

  • An example using the inpainting technique can be found at opencv_source_code/samples/cpp/inpaint.cpp

  • (Python) An example using the inpainting technique can be found at opencv_source_code/samples/python/inpaint.py

Python prototype (for reference only):

inpaint(src, inpaintMask, inpaintRadius, flags[, dst]) -> dst
Link to this function

inpaint(src, inpaintMask, inpaintRadius, flags, opts)

View Source
@spec inpaint(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  number(),
  integer(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Restores the selected region in an image using the region neighborhood.

Positional Arguments
  • src: Evision.Mat.t().

    Input 8-bit, 16-bit unsigned or 32-bit float 1-channel or 8-bit 3-channel image.

  • inpaintMask: Evision.Mat.t().

    Inpainting mask, 8-bit 1-channel image. Non-zero pixels indicate the area that needs to be inpainted.

  • inpaintRadius: double.

    Radius of a circular neighborhood of each point inpainted that is considered by the algorithm.

  • flags: int.

    Inpainting method that could be cv::INPAINT_NS or cv::INPAINT_TELEA

Return
  • dst: Evision.Mat.t().

    Output image with the same size and type as src .

The function reconstructs the selected image area from the pixel near the area boundary. The function may be used to remove dust and scratches from a scanned photo, or to remove undesirable objects from still images or video. See http://en.wikipedia.org/wiki/Inpainting for more details. Note:

  • An example using the inpainting technique can be found at opencv_source_code/samples/cpp/inpaint.cpp

  • (Python) An example using the inpainting technique can be found at opencv_source_code/samples/python/inpaint.py

Python prototype (for reference only):

inpaint(src, inpaintMask, inpaintRadius, flags[, dst]) -> dst
Link to this function

inRange(src, lowerb, upperb)

View Source

Checks if array elements lie between the elements of two other arrays.

Positional Arguments
  • src: Evision.Mat.t().

    first input array.

  • lowerb: Evision.Mat.t().

    inclusive lower boundary array or a scalar.

  • upperb: Evision.Mat.t().

    inclusive upper boundary array or a scalar.

Return
  • dst: Evision.Mat.t().

    output array of the same size as src and CV_8U type.

The function checks the range as follows:

  • For every element of a single-channel input array: \f[\texttt{dst} (I)= \texttt{lowerb} (I)_0 \leq \texttt{src} (I)_0 \leq \texttt{upperb} (I)_0\f]

  • For two-channel arrays: \f[\texttt{dst} (I)= \texttt{lowerb} (I)_0 \leq \texttt{src} (I)_0 \leq \texttt{upperb} (I)_0 \land \texttt{lowerb} (I)_1 \leq \texttt{src} (I)_1 \leq \texttt{upperb} (I)_1\f]

  • and so forth.

That is, dst (I) is set to 255 (all 1 -bits) if src (I) is within the specified 1D, 2D, 3D, ... box and 0 otherwise. When the lower and/or upper boundary parameters are scalars, the indexes (I) at lowerb and upperb in the above formulas should be omitted.

Python prototype (for reference only):

inRange(src, lowerb, upperb[, dst]) -> dst
Link to this function

inRange(src, lowerb, upperb, opts)

View Source
@spec inRange(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Checks if array elements lie between the elements of two other arrays.

Positional Arguments
  • src: Evision.Mat.t().

    first input array.

  • lowerb: Evision.Mat.t().

    inclusive lower boundary array or a scalar.

  • upperb: Evision.Mat.t().

    inclusive upper boundary array or a scalar.

Return
  • dst: Evision.Mat.t().

    output array of the same size as src and CV_8U type.

The function checks the range as follows:

  • For every element of a single-channel input array: \f[\texttt{dst} (I)= \texttt{lowerb} (I)_0 \leq \texttt{src} (I)_0 \leq \texttt{upperb} (I)_0\f]

  • For two-channel arrays: \f[\texttt{dst} (I)= \texttt{lowerb} (I)_0 \leq \texttt{src} (I)_0 \leq \texttt{upperb} (I)_0 \land \texttt{lowerb} (I)_1 \leq \texttt{src} (I)_1 \leq \texttt{upperb} (I)_1\f]

  • and so forth.

That is, dst (I) is set to 255 (all 1 -bits) if src (I) is within the specified 1D, 2D, 3D, ... box and 0 otherwise. When the lower and/or upper boundary parameters are scalars, the indexes (I) at lowerb and upperb in the above formulas should be omitted.

Python prototype (for reference only):

inRange(src, lowerb, upperb[, dst]) -> dst
Link to this function

insertChannel(src, dst, coi)

View Source
@spec insertChannel(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in(), integer()) ::
  Evision.Mat.t() | {:error, String.t()}

Inserts a single channel to dst (coi is 0-based index)

Positional Arguments
  • src: Evision.Mat.t().

    input array

  • coi: int.

    index of channel for insertion

Return
  • dst: Evision.Mat.t().

    output array

@sa mixChannels, merge

Python prototype (for reference only):

insertChannel(src, dst, coi) -> dst
@spec integral2(Evision.Mat.maybe_mat_in()) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

integral2

Positional Arguments
  • src: Evision.Mat.t()
Keyword Arguments
  • sdepth: int.
  • sqdepth: int.
Return
  • sum: Evision.Mat.t().
  • sqsum: Evision.Mat.t().

Has overloading in C++

Python prototype (for reference only):

integral2(src[, sum[, sqsum[, sdepth[, sqdepth]]]]) -> sum, sqsum
@spec integral2(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

integral2

Positional Arguments
  • src: Evision.Mat.t()
Keyword Arguments
  • sdepth: int.
  • sqdepth: int.
Return
  • sum: Evision.Mat.t().
  • sqsum: Evision.Mat.t().

Has overloading in C++

Python prototype (for reference only):

integral2(src[, sum[, sqsum[, sdepth[, sqdepth]]]]) -> sum, sqsum
@spec integral3(Evision.Mat.maybe_mat_in()) ::
  {Evision.Mat.t(), Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Calculates the integral of an image.

Positional Arguments
  • src: Evision.Mat.t().

    input image as \f$W \times H\f$, 8-bit or floating-point (32f or 64f).

Keyword Arguments
  • sdepth: int.

    desired depth of the integral and the tilted integral images, CV_32S, CV_32F, or CV_64F.

  • sqdepth: int.

    desired depth of the integral image of squared pixel values, CV_32F or CV_64F.

Return
  • sum: Evision.Mat.t().

    integral image as \f$(W+1)\times (H+1)\f$ , 32-bit integer or floating-point (32f or 64f).

  • sqsum: Evision.Mat.t().

    integral image for squared pixel values; it is \f$(W+1)\times (H+1)\f$, double-precision floating-point (64f) array.

  • tilted: Evision.Mat.t().

    integral for the image rotated by 45 degrees; it is \f$(W+1)\times (H+1)\f$ array with the same data type as sum.

The function calculates one or more integral images for the source image as follows: \f[\texttt{sum} (X,Y) = \sum _{x<X,y<Y} \texttt{image} (x,y)\f] \f[\texttt{sqsum} (X,Y) = \sum _{x<X,y<Y} \texttt{image} (x,y)^2\f] \f[\texttt{tilted} (X,Y) = \sum _{y<Y,abs(x-X+1) \leq Y-y-1} \texttt{image} (x,y)\f] Using these integral images, you can calculate sum, mean, and standard deviation over a specific up-right or rotated rectangular region of the image in a constant time, for example: \f[\sum _{x_1 \leq x < x_2, \, y_1 \leq y < y_2} \texttt{image} (x,y) = \texttt{sum} (x_2,y_2)- \texttt{sum} (x_1,y_2)- \texttt{sum} (x_2,y_1)+ \texttt{sum} (x_1,y_1)\f] It makes possible to do a fast blurring or fast block correlation with a variable window size, for example. In case of multi-channel images, sums for each channel are accumulated independently. As a practical example, the next figure shows the calculation of the integral of a straight rectangle Rect(4,4,3,2) and of a tilted rectangle Rect(5,1,2,3) . The selected pixels in the original image are shown, as well as the relative pixels in the integral images sum and tilted . integral calculation example

Python prototype (for reference only):

integral3(src[, sum[, sqsum[, tilted[, sdepth[, sqdepth]]]]]) -> sum, sqsum, tilted
@spec integral3(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  {Evision.Mat.t(), Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Calculates the integral of an image.

Positional Arguments
  • src: Evision.Mat.t().

    input image as \f$W \times H\f$, 8-bit or floating-point (32f or 64f).

Keyword Arguments
  • sdepth: int.

    desired depth of the integral and the tilted integral images, CV_32S, CV_32F, or CV_64F.

  • sqdepth: int.

    desired depth of the integral image of squared pixel values, CV_32F or CV_64F.

Return
  • sum: Evision.Mat.t().

    integral image as \f$(W+1)\times (H+1)\f$ , 32-bit integer or floating-point (32f or 64f).

  • sqsum: Evision.Mat.t().

    integral image for squared pixel values; it is \f$(W+1)\times (H+1)\f$, double-precision floating-point (64f) array.

  • tilted: Evision.Mat.t().

    integral for the image rotated by 45 degrees; it is \f$(W+1)\times (H+1)\f$ array with the same data type as sum.

The function calculates one or more integral images for the source image as follows: \f[\texttt{sum} (X,Y) = \sum _{x<X,y<Y} \texttt{image} (x,y)\f] \f[\texttt{sqsum} (X,Y) = \sum _{x<X,y<Y} \texttt{image} (x,y)^2\f] \f[\texttt{tilted} (X,Y) = \sum _{y<Y,abs(x-X+1) \leq Y-y-1} \texttt{image} (x,y)\f] Using these integral images, you can calculate sum, mean, and standard deviation over a specific up-right or rotated rectangular region of the image in a constant time, for example: \f[\sum _{x_1 \leq x < x_2, \, y_1 \leq y < y_2} \texttt{image} (x,y) = \texttt{sum} (x_2,y_2)- \texttt{sum} (x_1,y_2)- \texttt{sum} (x_2,y_1)+ \texttt{sum} (x_1,y_1)\f] It makes possible to do a fast blurring or fast block correlation with a variable window size, for example. In case of multi-channel images, sums for each channel are accumulated independently. As a practical example, the next figure shows the calculation of the integral of a straight rectangle Rect(4,4,3,2) and of a tilted rectangle Rect(5,1,2,3) . The selected pixels in the original image are shown, as well as the relative pixels in the integral images sum and tilted . integral calculation example

Python prototype (for reference only):

integral3(src[, sum[, sqsum[, tilted[, sdepth[, sqdepth]]]]]) -> sum, sqsum, tilted
@spec integral(Evision.Mat.maybe_mat_in()) :: Evision.Mat.t() | {:error, String.t()}

integral

Positional Arguments
  • src: Evision.Mat.t()
Keyword Arguments
  • sdepth: int.
Return
  • sum: Evision.Mat.t().

Has overloading in C++

Python prototype (for reference only):

integral(src[, sum[, sdepth]]) -> sum
@spec integral(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

integral

Positional Arguments
  • src: Evision.Mat.t()
Keyword Arguments
  • sdepth: int.
Return
  • sum: Evision.Mat.t().

Has overloading in C++

Python prototype (for reference only):

integral(src[, sum[, sdepth]]) -> sum
Link to this function

intersectConvexConvex(p1, p2)

View Source
@spec intersectConvexConvex(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in()) ::
  {number(), Evision.Mat.t()} | {:error, String.t()}

Finds intersection of two convex polygons

Positional Arguments
  • p1: Evision.Mat.t().

    First polygon

  • p2: Evision.Mat.t().

    Second polygon

Keyword Arguments
  • handleNested: bool.

    When true, an intersection is found if one of the polygons is fully enclosed in the other. When false, no intersection is found. If the polygons share a side or the vertex of one polygon lies on an edge of the other, they are not considered nested and an intersection will be found regardless of the value of handleNested.

Return
  • retval: float

  • p12: Evision.Mat.t().

    Output polygon describing the intersecting area

@returns Absolute value of area of intersecting polygon Note: intersectConvexConvex doesn't confirm that both polygons are convex and will return invalid results if they aren't.

Python prototype (for reference only):

intersectConvexConvex(p1, p2[, p12[, handleNested]]) -> retval, p12
Link to this function

intersectConvexConvex(p1, p2, opts)

View Source
@spec intersectConvexConvex(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: {number(), Evision.Mat.t()} | {:error, String.t()}

Finds intersection of two convex polygons

Positional Arguments
  • p1: Evision.Mat.t().

    First polygon

  • p2: Evision.Mat.t().

    Second polygon

Keyword Arguments
  • handleNested: bool.

    When true, an intersection is found if one of the polygons is fully enclosed in the other. When false, no intersection is found. If the polygons share a side or the vertex of one polygon lies on an edge of the other, they are not considered nested and an intersection will be found regardless of the value of handleNested.

Return
  • retval: float

  • p12: Evision.Mat.t().

    Output polygon describing the intersecting area

@returns Absolute value of area of intersecting polygon Note: intersectConvexConvex doesn't confirm that both polygons are convex and will return invalid results if they aren't.

Python prototype (for reference only):

intersectConvexConvex(p1, p2[, p12[, handleNested]]) -> retval, p12
@spec invert(Evision.Mat.maybe_mat_in()) ::
  {number(), Evision.Mat.t()} | {:error, String.t()}

Finds the inverse or pseudo-inverse of a matrix.

Positional Arguments
  • src: Evision.Mat.t().

    input floating-point M x N matrix.

Keyword Arguments
  • flags: int.

    inversion method (cv::DecompTypes)

Return
  • retval: double

  • dst: Evision.Mat.t().

    output matrix of N x M size and the same type as src.

The function cv::invert inverts the matrix src and stores the result in dst . When the matrix src is singular or non-square, the function calculates the pseudo-inverse matrix (the dst matrix) so that norm(src*dst - I) is minimal, where I is an identity matrix. In case of the #DECOMP_LU method, the function returns non-zero value if the inverse has been successfully calculated and 0 if src is singular. In case of the #DECOMP_SVD method, the function returns the inverse condition number of src (the ratio of the smallest singular value to the largest singular value) and 0 if src is singular. The SVD method calculates a pseudo-inverse matrix if src is singular. Similarly to #DECOMP_LU, the method #DECOMP_CHOLESKY works only with non-singular square matrices that should also be symmetrical and positively defined. In this case, the function stores the inverted matrix in dst and returns non-zero. Otherwise, it returns 0. @sa solve, SVD

Python prototype (for reference only):

invert(src[, dst[, flags]]) -> retval, dst
@spec invert(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  {number(), Evision.Mat.t()} | {:error, String.t()}

Finds the inverse or pseudo-inverse of a matrix.

Positional Arguments
  • src: Evision.Mat.t().

    input floating-point M x N matrix.

Keyword Arguments
  • flags: int.

    inversion method (cv::DecompTypes)

Return
  • retval: double

  • dst: Evision.Mat.t().

    output matrix of N x M size and the same type as src.

The function cv::invert inverts the matrix src and stores the result in dst . When the matrix src is singular or non-square, the function calculates the pseudo-inverse matrix (the dst matrix) so that norm(src*dst - I) is minimal, where I is an identity matrix. In case of the #DECOMP_LU method, the function returns non-zero value if the inverse has been successfully calculated and 0 if src is singular. In case of the #DECOMP_SVD method, the function returns the inverse condition number of src (the ratio of the smallest singular value to the largest singular value) and 0 if src is singular. The SVD method calculates a pseudo-inverse matrix if src is singular. Similarly to #DECOMP_LU, the method #DECOMP_CHOLESKY works only with non-singular square matrices that should also be symmetrical and positively defined. In this case, the function stores the inverted matrix in dst and returns non-zero. Otherwise, it returns 0. @sa solve, SVD

Python prototype (for reference only):

invert(src[, dst[, flags]]) -> retval, dst
Link to this function

invertAffineTransform(m)

View Source
@spec invertAffineTransform(Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

Inverts an affine transformation.

Positional Arguments
  • m: Evision.Mat.t().

    Original affine transformation.

Return
  • iM: Evision.Mat.t().

    Output reverse affine transformation.

The function computes an inverse affine transformation represented by \f$2 \times 3\f$ matrix M: \f[\begin{bmatrix} a_{11} & a_{12} & b_1 \\ a_{21} & a_{22} & b_2 \end{bmatrix}\f] The result is also a \f$2 \times 3\f$ matrix of the same type as M.

Python prototype (for reference only):

invertAffineTransform(M[, iM]) -> iM
Link to this function

invertAffineTransform(m, opts)

View Source
@spec invertAffineTransform(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Inverts an affine transformation.

Positional Arguments
  • m: Evision.Mat.t().

    Original affine transformation.

Return
  • iM: Evision.Mat.t().

    Output reverse affine transformation.

The function computes an inverse affine transformation represented by \f$2 \times 3\f$ matrix M: \f[\begin{bmatrix} a_{11} & a_{12} & b_1 \\ a_{21} & a_{22} & b_2 \end{bmatrix}\f] The result is also a \f$2 \times 3\f$ matrix of the same type as M.

Python prototype (for reference only):

invertAffineTransform(M[, iM]) -> iM
Link to this function

isContourConvex(contour)

View Source
@spec isContourConvex(Evision.Mat.maybe_mat_in()) :: boolean() | {:error, String.t()}

Tests a contour convexity.

Positional Arguments
  • contour: Evision.Mat.t().

    Input vector of 2D points, stored in std::vector\<> or Mat

Return
  • retval: bool

The function tests whether the input contour is convex or not. The contour must be simple, that is, without self-intersections. Otherwise, the function output is undefined.

Python prototype (for reference only):

isContourConvex(contour) -> retval
Link to this function

kmeans(data, k, bestLabels, criteria, attempts, flags)

View Source

Finds centers of clusters and groups input samples around the clusters.

Positional Arguments
  • data: Evision.Mat.t().

    Data for clustering. An array of N-Dimensional points with float coordinates is needed. Examples of this array can be:

    • Mat points(count, 2, CV_32F);
    • Mat points(count, 1, CV_32FC2);
    • Mat points(1, count, CV_32FC2);
    • std::vector\<cv::Point2f> points(sampleCount);
  • k: int.

    Number of clusters to split the set by.

  • criteria: TermCriteria.

    The algorithm termination criteria, that is, the maximum number of iterations and/or the desired accuracy. The accuracy is specified as criteria.epsilon. As soon as each of the cluster centers moves by less than criteria.epsilon on some iteration, the algorithm stops.

  • attempts: int.

    Flag to specify the number of times the algorithm is executed using different initial labellings. The algorithm returns the labels that yield the best compactness (see the last function parameter).

  • flags: int.

    Flag that can take values of cv::KmeansFlags

Return
  • retval: double

  • bestLabels: Evision.Mat.t().

    Input/output integer array that stores the cluster indices for every sample.

  • centers: Evision.Mat.t().

    Output matrix of the cluster centers, one row per each cluster center.

The function kmeans implements a k-means algorithm that finds the centers of cluster_count clusters and groups the input samples around the clusters. As an output, \f$\texttt{bestLabels}_i\f$ contains a 0-based cluster index for the sample stored in the \f$i^{th}\f$ row of the samples matrix. Note:

  • (Python) An example on K-means clustering can be found at opencv_source_code/samples/python/kmeans.py

@return The function returns the compactness measure that is computed as \f[\sum _i \| \texttt{samples} _i - \texttt{centers} _{ \texttt{labels} _i} \| ^2\f] after every attempt. The best (minimum) value is chosen and the corresponding labels and the compactness value are returned by the function. Basically, you can use only the core of the function, set the number of attempts to 1, initialize labels each time using a custom algorithm, pass them with the ( flags = #KMEANS_USE_INITIAL_LABELS ) flag, and then choose the best (most-compact) clustering.

Python prototype (for reference only):

kmeans(data, K, bestLabels, criteria, attempts, flags[, centers]) -> retval, bestLabels, centers
Link to this function

kmeans(data, k, bestLabels, criteria, attempts, flags, opts)

View Source
@spec kmeans(
  Evision.Mat.maybe_mat_in(),
  integer(),
  Evision.Mat.maybe_mat_in(),
  {integer(), integer(), number()},
  integer(),
  integer(),
  [{atom(), term()}, ...] | nil
) :: {number(), Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Finds centers of clusters and groups input samples around the clusters.

Positional Arguments
  • data: Evision.Mat.t().

    Data for clustering. An array of N-Dimensional points with float coordinates is needed. Examples of this array can be:

    • Mat points(count, 2, CV_32F);
    • Mat points(count, 1, CV_32FC2);
    • Mat points(1, count, CV_32FC2);
    • std::vector\<cv::Point2f> points(sampleCount);
  • k: int.

    Number of clusters to split the set by.

  • criteria: TermCriteria.

    The algorithm termination criteria, that is, the maximum number of iterations and/or the desired accuracy. The accuracy is specified as criteria.epsilon. As soon as each of the cluster centers moves by less than criteria.epsilon on some iteration, the algorithm stops.

  • attempts: int.

    Flag to specify the number of times the algorithm is executed using different initial labellings. The algorithm returns the labels that yield the best compactness (see the last function parameter).

  • flags: int.

    Flag that can take values of cv::KmeansFlags

Return
  • retval: double

  • bestLabels: Evision.Mat.t().

    Input/output integer array that stores the cluster indices for every sample.

  • centers: Evision.Mat.t().

    Output matrix of the cluster centers, one row per each cluster center.

The function kmeans implements a k-means algorithm that finds the centers of cluster_count clusters and groups the input samples around the clusters. As an output, \f$\texttt{bestLabels}_i\f$ contains a 0-based cluster index for the sample stored in the \f$i^{th}\f$ row of the samples matrix. Note:

  • (Python) An example on K-means clustering can be found at opencv_source_code/samples/python/kmeans.py

@return The function returns the compactness measure that is computed as \f[\sum _i \| \texttt{samples} _i - \texttt{centers} _{ \texttt{labels} _i} \| ^2\f] after every attempt. The best (minimum) value is chosen and the corresponding labels and the compactness value are returned by the function. Basically, you can use only the core of the function, set the number of attempts to 1, initialize labels each time using a custom algorithm, pass them with the ( flags = #KMEANS_USE_INITIAL_LABELS ) flag, and then choose the best (most-compact) clustering.

Python prototype (for reference only):

kmeans(data, K, bestLabels, criteria, attempts, flags[, centers]) -> retval, bestLabels, centers
@spec laplacian(Evision.Mat.maybe_mat_in(), integer()) ::
  Evision.Mat.t() | {:error, String.t()}

Calculates the Laplacian of an image.

Positional Arguments
  • src: Evision.Mat.t().

    Source image.

  • ddepth: int.

    Desired depth of the destination image, see @ref filter_depths "combinations".

Keyword Arguments
  • ksize: int.

    Aperture size used to compute the second-derivative filters. See #getDerivKernels for details. The size must be positive and odd.

  • scale: double.

    Optional scale factor for the computed Laplacian values. By default, no scaling is applied. See #getDerivKernels for details.

  • delta: double.

    Optional delta value that is added to the results prior to storing them in dst .

  • borderType: int.

    Pixel extrapolation method, see #BorderTypes. #BORDER_WRAP is not supported.

Return
  • dst: Evision.Mat.t().

    Destination image of the same size and the same number of channels as src .

The function calculates the Laplacian of the source image by adding up the second x and y derivatives calculated using the Sobel operator: \f[\texttt{dst} = \Delta \texttt{src} = \frac{\partial^2 \texttt{src}}{\partial x^2} + \frac{\partial^2 \texttt{src}}{\partial y^2}\f] This is done when ksize > 1. When ksize == 1, the Laplacian is computed by filtering the image with the following \f$3 \times 3\f$ aperture: \f[\vecthreethree {0}{1}{0}{1}{-4}{1}{0}{1}{0}\f] @sa Sobel, Scharr

Python prototype (for reference only):

Laplacian(src, ddepth[, dst[, ksize[, scale[, delta[, borderType]]]]]) -> dst
Link to this function

laplacian(src, ddepth, opts)

View Source
@spec laplacian(Evision.Mat.maybe_mat_in(), integer(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Calculates the Laplacian of an image.

Positional Arguments
  • src: Evision.Mat.t().

    Source image.

  • ddepth: int.

    Desired depth of the destination image, see @ref filter_depths "combinations".

Keyword Arguments
  • ksize: int.

    Aperture size used to compute the second-derivative filters. See #getDerivKernels for details. The size must be positive and odd.

  • scale: double.

    Optional scale factor for the computed Laplacian values. By default, no scaling is applied. See #getDerivKernels for details.

  • delta: double.

    Optional delta value that is added to the results prior to storing them in dst .

  • borderType: int.

    Pixel extrapolation method, see #BorderTypes. #BORDER_WRAP is not supported.

Return
  • dst: Evision.Mat.t().

    Destination image of the same size and the same number of channels as src .

The function calculates the Laplacian of the source image by adding up the second x and y derivatives calculated using the Sobel operator: \f[\texttt{dst} = \Delta \texttt{src} = \frac{\partial^2 \texttt{src}}{\partial x^2} + \frac{\partial^2 \texttt{src}}{\partial y^2}\f] This is done when ksize > 1. When ksize == 1, the Laplacian is computed by filtering the image with the following \f$3 \times 3\f$ aperture: \f[\vecthreethree {0}{1}{0}{1}{-4}{1}{0}{1}{0}\f] @sa Sobel, Scharr

Python prototype (for reference only):

Laplacian(src, ddepth[, dst[, ksize[, scale[, delta[, borderType]]]]]) -> dst
Link to this function

line(img, pt1, pt2, color)

View Source
@spec line(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  {number(), number()},
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()}
) :: Evision.Mat.t() | {:error, String.t()}

Draws a line segment connecting two points.

Positional Arguments
  • pt1: Point.

    First point of the line segment.

  • pt2: Point.

    Second point of the line segment.

  • color: Scalar.

    Line color.

Keyword Arguments
  • thickness: int.

    Line thickness.

  • lineType: int.

    Type of the line. See #LineTypes.

  • shift: int.

    Number of fractional bits in the point coordinates.

Return
  • img: Evision.Mat.t().

    Image.

The function line draws the line segment between pt1 and pt2 points in the image. The line is clipped by the image boundaries. For non-antialiased lines with integer coordinates, the 8-connected or 4-connected Bresenham algorithm is used. Thick lines are drawn with rounding endings. Antialiased lines are drawn using Gaussian filtering.

Python prototype (for reference only):

line(img, pt1, pt2, color[, thickness[, lineType[, shift]]]) -> img
Link to this function

line(img, pt1, pt2, color, opts)

View Source
@spec line(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  {number(), number()},
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()},
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Draws a line segment connecting two points.

Positional Arguments
  • pt1: Point.

    First point of the line segment.

  • pt2: Point.

    Second point of the line segment.

  • color: Scalar.

    Line color.

Keyword Arguments
  • thickness: int.

    Line thickness.

  • lineType: int.

    Type of the line. See #LineTypes.

  • shift: int.

    Number of fractional bits in the point coordinates.

Return
  • img: Evision.Mat.t().

    Image.

The function line draws the line segment between pt1 and pt2 points in the image. The line is clipped by the image boundaries. For non-antialiased lines with integer coordinates, the 8-connected or 4-connected Bresenham algorithm is used. Thick lines are drawn with rounding endings. Antialiased lines are drawn using Gaussian filtering.

Python prototype (for reference only):

line(img, pt1, pt2, color[, thickness[, lineType[, shift]]]) -> img
Link to this function

linearPolar(src, center, maxRadius, flags)

View Source
@spec linearPolar(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  number(),
  integer()
) ::
  Evision.Mat.t() | {:error, String.t()}

Remaps an image to polar coordinates space.

Positional Arguments
  • src: Evision.Mat.t().

    Source image

  • center: Point2f.

    The transformation center;

  • maxRadius: double.

    The radius of the bounding circle to transform. It determines the inverse magnitude scale parameter too.

  • flags: int.

    A combination of interpolation methods, see #InterpolationFlags

Return
  • dst: Evision.Mat.t().

    Destination image. It will have same size and type as src.

@deprecated This function produces same result as cv::warpPolar(src, dst, src.size(), center, maxRadius, flags) @internal Transform the source image using the following transformation (See @ref polar_remaps_reference_image "Polar remaps reference image c)"): \f[\begin{array}{l} dst( \rho , \phi ) = src(x,y) \\ dst.size() \leftarrow src.size() \end{array}\f] where \f[\begin{array}{l} I = (dx,dy) = (x - center.x,y - center.y) \\ \rho = Kmag \cdot \texttt{magnitude} (I) ,\\ \phi = angle \cdot \texttt{angle} (I) \end{array}\f] and \f[\begin{array}{l} Kx = src.cols / maxRadius \\ Ky = src.rows / 2\Pi \end{array}\f]

Note:

  • The function can not operate in-place.
  • To calculate magnitude and angle in degrees #cartToPolar is used internally thus angles are measured from 0 to 360 with accuracy about 0.3 degrees.

@sa cv::logPolar @endinternal

Python prototype (for reference only):

linearPolar(src, center, maxRadius, flags[, dst]) -> dst
Link to this function

linearPolar(src, center, maxRadius, flags, opts)

View Source
@spec linearPolar(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  number(),
  integer(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Remaps an image to polar coordinates space.

Positional Arguments
  • src: Evision.Mat.t().

    Source image

  • center: Point2f.

    The transformation center;

  • maxRadius: double.

    The radius of the bounding circle to transform. It determines the inverse magnitude scale parameter too.

  • flags: int.

    A combination of interpolation methods, see #InterpolationFlags

Return
  • dst: Evision.Mat.t().

    Destination image. It will have same size and type as src.

@deprecated This function produces same result as cv::warpPolar(src, dst, src.size(), center, maxRadius, flags) @internal Transform the source image using the following transformation (See @ref polar_remaps_reference_image "Polar remaps reference image c)"): \f[\begin{array}{l} dst( \rho , \phi ) = src(x,y) \\ dst.size() \leftarrow src.size() \end{array}\f] where \f[\begin{array}{l} I = (dx,dy) = (x - center.x,y - center.y) \\ \rho = Kmag \cdot \texttt{magnitude} (I) ,\\ \phi = angle \cdot \texttt{angle} (I) \end{array}\f] and \f[\begin{array}{l} Kx = src.cols / maxRadius \\ Ky = src.rows / 2\Pi \end{array}\f]

Note:

  • The function can not operate in-place.
  • To calculate magnitude and angle in degrees #cartToPolar is used internally thus angles are measured from 0 to 360 with accuracy about 0.3 degrees.

@sa cv::logPolar @endinternal

Python prototype (for reference only):

linearPolar(src, center, maxRadius, flags[, dst]) -> dst
@spec log(Evision.Mat.maybe_mat_in()) :: Evision.Mat.t() | {:error, String.t()}

Calculates the natural logarithm of every array element.

Positional Arguments
  • src: Evision.Mat.t().

    input array.

Return
  • dst: Evision.Mat.t().

    output array of the same size and type as src .

The function cv::log calculates the natural logarithm of every element of the input array: \f[\texttt{dst} (I) = \log (\texttt{src}(I)) \f] Output on zero, negative and special (NaN, Inf) values is undefined. @sa exp, cartToPolar, polarToCart, phase, pow, sqrt, magnitude

Python prototype (for reference only):

log(src[, dst]) -> dst
@spec log(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Calculates the natural logarithm of every array element.

Positional Arguments
  • src: Evision.Mat.t().

    input array.

Return
  • dst: Evision.Mat.t().

    output array of the same size and type as src .

The function cv::log calculates the natural logarithm of every element of the input array: \f[\texttt{dst} (I) = \log (\texttt{src}(I)) \f] Output on zero, negative and special (NaN, Inf) values is undefined. @sa exp, cartToPolar, polarToCart, phase, pow, sqrt, magnitude

Python prototype (for reference only):

log(src[, dst]) -> dst
Link to this function

logPolar(src, center, m, flags)

View Source
@spec logPolar(Evision.Mat.maybe_mat_in(), {number(), number()}, number(), integer()) ::
  Evision.Mat.t() | {:error, String.t()}

Remaps an image to semilog-polar coordinates space.

Positional Arguments
  • src: Evision.Mat.t().

    Source image

  • center: Point2f.

    The transformation center; where the output precision is maximal

  • m: double.

    Magnitude scale parameter. It determines the radius of the bounding circle to transform too.

  • flags: int.

    A combination of interpolation methods, see #InterpolationFlags

Return
  • dst: Evision.Mat.t().

    Destination image. It will have same size and type as src.

@deprecated This function produces same result as cv::warpPolar(src, dst, src.size(), center, maxRadius, flags+WARP_POLAR_LOG); @internal Transform the source image using the following transformation (See @ref polar_remaps_reference_image "Polar remaps reference image d)"): \f[\begin{array}{l} dst( \rho , \phi ) = src(x,y) \\ dst.size() \leftarrow src.size() \end{array}\f] where \f[\begin{array}{l} I = (dx,dy) = (x - center.x,y - center.y) \\ \rho = M \cdot log_e(\texttt{magnitude} (I)) ,\\ \phi = Kangle \cdot \texttt{angle} (I) \\ \end{array}\f] and \f[\begin{array}{l} M = src.cols / log_e(maxRadius) \\ Kangle = src.rows / 2\Pi \\ \end{array}\f] The function emulates the human "foveal" vision and can be used for fast scale and rotation-invariant template matching, for object tracking and so forth.

Note:

  • The function can not operate in-place.
  • To calculate magnitude and angle in degrees #cartToPolar is used internally thus angles are measured from 0 to 360 with accuracy about 0.3 degrees.

@sa cv::linearPolar @endinternal

Python prototype (for reference only):

logPolar(src, center, M, flags[, dst]) -> dst
Link to this function

logPolar(src, center, m, flags, opts)

View Source
@spec logPolar(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  number(),
  integer(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Remaps an image to semilog-polar coordinates space.

Positional Arguments
  • src: Evision.Mat.t().

    Source image

  • center: Point2f.

    The transformation center; where the output precision is maximal

  • m: double.

    Magnitude scale parameter. It determines the radius of the bounding circle to transform too.

  • flags: int.

    A combination of interpolation methods, see #InterpolationFlags

Return
  • dst: Evision.Mat.t().

    Destination image. It will have same size and type as src.

@deprecated This function produces same result as cv::warpPolar(src, dst, src.size(), center, maxRadius, flags+WARP_POLAR_LOG); @internal Transform the source image using the following transformation (See @ref polar_remaps_reference_image "Polar remaps reference image d)"): \f[\begin{array}{l} dst( \rho , \phi ) = src(x,y) \\ dst.size() \leftarrow src.size() \end{array}\f] where \f[\begin{array}{l} I = (dx,dy) = (x - center.x,y - center.y) \\ \rho = M \cdot log_e(\texttt{magnitude} (I)) ,\\ \phi = Kangle \cdot \texttt{angle} (I) \\ \end{array}\f] and \f[\begin{array}{l} M = src.cols / log_e(maxRadius) \\ Kangle = src.rows / 2\Pi \\ \end{array}\f] The function emulates the human "foveal" vision and can be used for fast scale and rotation-invariant template matching, for object tracking and so forth.

Note:

  • The function can not operate in-place.
  • To calculate magnitude and angle in degrees #cartToPolar is used internally thus angles are measured from 0 to 360 with accuracy about 0.3 degrees.

@sa cv::linearPolar @endinternal

Python prototype (for reference only):

logPolar(src, center, M, flags[, dst]) -> dst

Performs a look-up table transform of an array.

Positional Arguments
  • src: Evision.Mat.t().

    input array of 8-bit elements.

  • lut: Evision.Mat.t().

    look-up table of 256 elements; in case of multi-channel input array, the table should either have a single channel (in this case the same table is used for all channels) or the same number of channels as in the input array.

Return
  • dst: Evision.Mat.t().

    output array of the same size and number of channels as src, and the same depth as lut.

The function LUT fills the output array with values from the look-up table. Indices of the entries are taken from the input array. That is, the function processes each element of src as follows: \f[\texttt{dst} (I) \leftarrow \texttt{lut(src(I) + d)}\f] where \f[d = \fork{0}{if (\texttt{src}) has depth (\texttt{CV_8U})}{128}{if (\texttt{src}) has depth (\texttt{CV_8S})}\f] @sa convertScaleAbs, Mat::convertTo

Python prototype (for reference only):

LUT(src, lut[, dst]) -> dst
@spec lut(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Performs a look-up table transform of an array.

Positional Arguments
  • src: Evision.Mat.t().

    input array of 8-bit elements.

  • lut: Evision.Mat.t().

    look-up table of 256 elements; in case of multi-channel input array, the table should either have a single channel (in this case the same table is used for all channels) or the same number of channels as in the input array.

Return
  • dst: Evision.Mat.t().

    output array of the same size and number of channels as src, and the same depth as lut.

The function LUT fills the output array with values from the look-up table. Indices of the entries are taken from the input array. That is, the function processes each element of src as follows: \f[\texttt{dst} (I) \leftarrow \texttt{lut(src(I) + d)}\f] where \f[d = \fork{0}{if (\texttt{src}) has depth (\texttt{CV_8U})}{128}{if (\texttt{src}) has depth (\texttt{CV_8S})}\f] @sa convertScaleAbs, Mat::convertTo

Python prototype (for reference only):

LUT(src, lut[, dst]) -> dst
@spec magnitude(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

Calculates the magnitude of 2D vectors.

Positional Arguments
  • x: Evision.Mat.t().

    floating-point array of x-coordinates of the vectors.

  • y: Evision.Mat.t().

    floating-point array of y-coordinates of the vectors; it must have the same size as x.

Return
  • magnitude: Evision.Mat.t().

    output array of the same size and type as x.

The function cv::magnitude calculates the magnitude of 2D vectors formed from the corresponding elements of x and y arrays: \f[\texttt{dst} (I) = \sqrt{\texttt{x}(I)^2 + \texttt{y}(I)^2}\f] @sa cartToPolar, polarToCart, phase, sqrt

Python prototype (for reference only):

magnitude(x, y[, magnitude]) -> magnitude
@spec magnitude(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Calculates the magnitude of 2D vectors.

Positional Arguments
  • x: Evision.Mat.t().

    floating-point array of x-coordinates of the vectors.

  • y: Evision.Mat.t().

    floating-point array of y-coordinates of the vectors; it must have the same size as x.

Return
  • magnitude: Evision.Mat.t().

    output array of the same size and type as x.

The function cv::magnitude calculates the magnitude of 2D vectors formed from the corresponding elements of x and y arrays: \f[\texttt{dst} (I) = \sqrt{\texttt{x}(I)^2 + \texttt{y}(I)^2}\f] @sa cartToPolar, polarToCart, phase, sqrt

Python prototype (for reference only):

magnitude(x, y[, magnitude]) -> magnitude
Link to this function

mahalanobis(v1, v2, icovar)

View Source

Calculates the Mahalanobis distance between two vectors.

Positional Arguments
  • v1: Evision.Mat.t().

    first 1D input vector.

  • v2: Evision.Mat.t().

    second 1D input vector.

  • icovar: Evision.Mat.t().

    inverse covariance matrix.

Return
  • retval: double

The function cv::Mahalanobis calculates and returns the weighted distance between two vectors: \f[d( \texttt{vec1} , \texttt{vec2} )= \sqrt{\sum_{i,j}{\texttt{icovar(i,j)}\cdot(\texttt{vec1}(I)-\texttt{vec2}(I))\cdot(\texttt{vec1(j)}-\texttt{vec2(j)})} }\f] The covariance matrix may be calculated using the #calcCovarMatrix function and then inverted using the invert function (preferably using the #DECOMP_SVD method, as the most accurate).

Python prototype (for reference only):

Mahalanobis(v1, v2, icovar) -> retval
Link to this function

matchShapes(contour1, contour2, method, parameter)

View Source
@spec matchShapes(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  integer(),
  number()
) ::
  number() | {:error, String.t()}

Compares two shapes.

Positional Arguments
  • contour1: Evision.Mat.t().

    First contour or grayscale image.

  • contour2: Evision.Mat.t().

    Second contour or grayscale image.

  • method: int.

    Comparison method, see #ShapeMatchModes

  • parameter: double.

    Method-specific parameter (not supported now).

Return
  • retval: double

The function compares two shapes. All three implemented methods use the Hu invariants (see #HuMoments)

Python prototype (for reference only):

matchShapes(contour1, contour2, method, parameter) -> retval
Link to this function

matchTemplate(image, templ, method)

View Source
@spec matchTemplate(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in(), integer()) ::
  Evision.Mat.t() | {:error, String.t()}

Compares a template against overlapped image regions.

Positional Arguments
  • image: Evision.Mat.t().

    Image where the search is running. It must be 8-bit or 32-bit floating-point.

  • templ: Evision.Mat.t().

    Searched template. It must be not greater than the source image and have the same data type.

  • method: int.

    Parameter specifying the comparison method, see #TemplateMatchModes

Keyword Arguments
  • mask: Evision.Mat.t().

    Optional mask. It must have the same size as templ. It must either have the same number of channels as template or only one channel, which is then used for all template and image channels. If the data type is #CV_8U, the mask is interpreted as a binary mask, meaning only elements where mask is nonzero are used and are kept unchanged independent of the actual mask value (weight equals 1). For data tpye #CV_32F, the mask values are used as weights. The exact formulas are documented in #TemplateMatchModes.

Return
  • result: Evision.Mat.t().

    Map of comparison results. It must be single-channel 32-bit floating-point. If image is \f$W \times H\f$ and templ is \f$w \times h\f$ , then result is \f$(W-w+1) \times (H-h+1)\f$ .

The function slides through image , compares the overlapped patches of size \f$w \times h\f$ against templ using the specified method and stores the comparison results in result . #TemplateMatchModes describes the formulae for the available comparison methods ( \f$I\f$ denotes image, \f$T\f$ template, \f$R\f$ result, \f$M\f$ the optional mask ). The summation is done over template and/or the image patch: \f$x' = 0...w-1, y' = 0...h-1\f$ After the function finishes the comparison, the best matches can be found as global minimums (when #TM_SQDIFF was used) or maximums (when #TM_CCORR or #TM_CCOEFF was used) using the #minMaxLoc function. In case of a color image, template summation in the numerator and each sum in the denominator is done over all of the channels and separate mean values are used for each channel. That is, the function can take a color template and a color image. The result will still be a single-channel image, which is easier to analyze.

Python prototype (for reference only):

matchTemplate(image, templ, method[, result[, mask]]) -> result
Link to this function

matchTemplate(image, templ, method, opts)

View Source
@spec matchTemplate(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  integer(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Compares a template against overlapped image regions.

Positional Arguments
  • image: Evision.Mat.t().

    Image where the search is running. It must be 8-bit or 32-bit floating-point.

  • templ: Evision.Mat.t().

    Searched template. It must be not greater than the source image and have the same data type.

  • method: int.

    Parameter specifying the comparison method, see #TemplateMatchModes

Keyword Arguments
  • mask: Evision.Mat.t().

    Optional mask. It must have the same size as templ. It must either have the same number of channels as template or only one channel, which is then used for all template and image channels. If the data type is #CV_8U, the mask is interpreted as a binary mask, meaning only elements where mask is nonzero are used and are kept unchanged independent of the actual mask value (weight equals 1). For data tpye #CV_32F, the mask values are used as weights. The exact formulas are documented in #TemplateMatchModes.

Return
  • result: Evision.Mat.t().

    Map of comparison results. It must be single-channel 32-bit floating-point. If image is \f$W \times H\f$ and templ is \f$w \times h\f$ , then result is \f$(W-w+1) \times (H-h+1)\f$ .

The function slides through image , compares the overlapped patches of size \f$w \times h\f$ against templ using the specified method and stores the comparison results in result . #TemplateMatchModes describes the formulae for the available comparison methods ( \f$I\f$ denotes image, \f$T\f$ template, \f$R\f$ result, \f$M\f$ the optional mask ). The summation is done over template and/or the image patch: \f$x' = 0...w-1, y' = 0...h-1\f$ After the function finishes the comparison, the best matches can be found as global minimums (when #TM_SQDIFF was used) or maximums (when #TM_CCORR or #TM_CCOEFF was used) using the #minMaxLoc function. In case of a color image, template summation in the numerator and each sum in the denominator is done over all of the channels and separate mean values are used for each channel. That is, the function can take a color template and a color image. The result will still be a single-channel image, which is easier to analyze.

Python prototype (for reference only):

matchTemplate(image, templ, method[, result[, mask]]) -> result

Computes partial derivatives of the matrix product for each multiplied matrix.

Positional Arguments
  • a: Evision.Mat.t().

    First multiplied matrix.

  • b: Evision.Mat.t().

    Second multiplied matrix.

Return
  • dABdA: Evision.Mat.t().

    First output derivative matrix d(A*B)/dA of size \f$\texttt{A.rowsB.cols} \times {A.rowsA.cols}\f$ .

  • dABdB: Evision.Mat.t().

    Second output derivative matrix d(A*B)/dB of size \f$\texttt{A.rowsB.cols} \times {B.rowsB.cols}\f$ .

The function computes partial derivatives of the elements of the matrix product \f$A*B\f$ with regard to the elements of each of the two input matrices. The function is used to compute the Jacobian matrices in #stereoCalibrate but can also be used in any other similar optimization function.

Python prototype (for reference only):

matMulDeriv(A, B[, dABdA[, dABdB]]) -> dABdA, dABdB
@spec matMulDeriv(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Computes partial derivatives of the matrix product for each multiplied matrix.

Positional Arguments
  • a: Evision.Mat.t().

    First multiplied matrix.

  • b: Evision.Mat.t().

    Second multiplied matrix.

Return
  • dABdA: Evision.Mat.t().

    First output derivative matrix d(A*B)/dA of size \f$\texttt{A.rowsB.cols} \times {A.rowsA.cols}\f$ .

  • dABdB: Evision.Mat.t().

    Second output derivative matrix d(A*B)/dB of size \f$\texttt{A.rowsB.cols} \times {B.rowsB.cols}\f$ .

The function computes partial derivatives of the elements of the matrix product \f$A*B\f$ with regard to the elements of each of the two input matrices. The function is used to compute the Jacobian matrices in #stereoCalibrate but can also be used in any other similar optimization function.

Python prototype (for reference only):

matMulDeriv(A, B[, dABdA[, dABdB]]) -> dABdA, dABdB

Calculates per-element maximum of two arrays or an array and a scalar.

Positional Arguments
  • src1: Evision.Mat.t().

    first input array.

  • src2: Evision.Mat.t().

    second input array of the same size and type as src1 .

Return
  • dst: Evision.Mat.t().

    output array of the same size and type as src1.

The function cv::max calculates the per-element maximum of two arrays: \f[\texttt{dst} (I)= \max ( \texttt{src1} (I), \texttt{src2} (I))\f] or array and a scalar: \f[\texttt{dst} (I)= \max ( \texttt{src1} (I), \texttt{value} )\f] @sa min, compare, inRange, minMaxLoc, @ref MatrixExpressions

Python prototype (for reference only):

max(src1, src2[, dst]) -> dst
@spec max(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Calculates per-element maximum of two arrays or an array and a scalar.

Positional Arguments
  • src1: Evision.Mat.t().

    first input array.

  • src2: Evision.Mat.t().

    second input array of the same size and type as src1 .

Return
  • dst: Evision.Mat.t().

    output array of the same size and type as src1.

The function cv::max calculates the per-element maximum of two arrays: \f[\texttt{dst} (I)= \max ( \texttt{src1} (I), \texttt{src2} (I))\f] or array and a scalar: \f[\texttt{dst} (I)= \max ( \texttt{src1} (I), \texttt{value} )\f] @sa min, compare, inRange, minMaxLoc, @ref MatrixExpressions

Python prototype (for reference only):

max(src1, src2[, dst]) -> dst
@spec mean(Evision.Mat.maybe_mat_in()) ::
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()}
  | {:error, String.t()}

Calculates an average (mean) of array elements.

Positional Arguments
  • src: Evision.Mat.t().

    input array that should have from 1 to 4 channels so that the result can be stored in Scalar_ .

Keyword Arguments
  • mask: Evision.Mat.t().

    optional operation mask.

Return
  • retval: Scalar

The function cv::mean calculates the mean value M of array elements, independently for each channel, and return it: \f[\begin{array}{l} N = \sum _{I: \; \texttt{mask} (I) \ne 0} 1 \\ M_c = \left ( \sum _{I: \; \texttt{mask} (I) \ne 0}{ \texttt{mtx} (I)_c} \right )/N \end{array}\f] When all the mask elements are 0's, the function returns Scalar::all(0) @sa countNonZero, meanStdDev, norm, minMaxLoc

Python prototype (for reference only):

mean(src[, mask]) -> retval
@spec mean(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()}
  | {:error, String.t()}

Calculates an average (mean) of array elements.

Positional Arguments
  • src: Evision.Mat.t().

    input array that should have from 1 to 4 channels so that the result can be stored in Scalar_ .

Keyword Arguments
  • mask: Evision.Mat.t().

    optional operation mask.

Return
  • retval: Scalar

The function cv::mean calculates the mean value M of array elements, independently for each channel, and return it: \f[\begin{array}{l} N = \sum _{I: \; \texttt{mask} (I) \ne 0} 1 \\ M_c = \left ( \sum _{I: \; \texttt{mask} (I) \ne 0}{ \texttt{mtx} (I)_c} \right )/N \end{array}\f] When all the mask elements are 0's, the function returns Scalar::all(0) @sa countNonZero, meanStdDev, norm, minMaxLoc

Python prototype (for reference only):

mean(src[, mask]) -> retval
Link to this function

meanShift(probImage, window, criteria)

View Source
@spec meanShift(
  Evision.Mat.maybe_mat_in(),
  {number(), number(), number(), number()},
  {integer(), integer(), number()}
) ::
  {integer(), {number(), number(), number(), number()}} | {:error, String.t()}

Finds an object on a back projection image.

Positional Arguments
  • probImage: Evision.Mat.t().

    Back projection of the object histogram. See calcBackProject for details.

  • criteria: TermCriteria.

    Stop criteria for the iterative search algorithm. returns : Number of iterations CAMSHIFT took to converge. The function implements the iterative object search algorithm. It takes the input back projection of an object and the initial position. The mass center in window of the back projection image is computed and the search window center shifts to the mass center. The procedure is repeated until the specified number of iterations criteria.maxCount is done or until the window center shifts by less than criteria.epsilon. The algorithm is used inside CamShift and, unlike CamShift , the search window size or orientation do not change during the search. You can simply pass the output of calcBackProject to this function. But better results can be obtained if you pre-filter the back projection and remove the noise. For example, you can do this by retrieving connected components with findContours , throwing away contours with small area ( contourArea ), and rendering the remaining contours with drawContours.

Return
  • retval: int

  • window: Rect.

    Initial search window.

Python prototype (for reference only):

meanShift(probImage, window, criteria) -> retval, window
@spec meanStdDev(Evision.Mat.maybe_mat_in()) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

meanStdDev

Positional Arguments
  • src: Evision.Mat.t().

    input array that should have from 1 to 4 channels so that the results can be stored in Scalar_ 's.

Keyword Arguments
  • mask: Evision.Mat.t().

    optional operation mask.

Return
  • mean: Evision.Mat.t().

    output parameter: calculated mean value.

  • stddev: Evision.Mat.t().

    output parameter: calculated standard deviation.

Calculates a mean and standard deviation of array elements. The function cv::meanStdDev calculates the mean and the standard deviation M of array elements independently for each channel and returns it via the output parameters: \f[\begin{array}{l} N = \sum _{I, \texttt{mask} (I) \ne 0} 1 \\ \texttt{mean} _c = \frac{\sum_{ I: \; \texttt{mask}(I) \ne 0} \texttt{src} (I)_c}{N} \\ \texttt{stddev} _c = \sqrt{\frac{\sum_{ I: \; \texttt{mask}(I) \ne 0} \left ( \texttt{src} (I)_c - \texttt{mean} _c \right )^2}{N}} \end{array}\f] When all the mask elements are 0's, the function returns mean=stddev=Scalar::all(0). Note: The calculated standard deviation is only the diagonal of the complete normalized covariance matrix. If the full matrix is needed, you can reshape the multi-channel array M x N to the single-channel array M*N x mtx.channels() (only possible when the matrix is continuous) and then pass the matrix to calcCovarMatrix . @sa countNonZero, mean, norm, minMaxLoc, calcCovarMatrix

Python prototype (for reference only):

meanStdDev(src[, mean[, stddev[, mask]]]) -> mean, stddev
@spec meanStdDev(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

meanStdDev

Positional Arguments
  • src: Evision.Mat.t().

    input array that should have from 1 to 4 channels so that the results can be stored in Scalar_ 's.

Keyword Arguments
  • mask: Evision.Mat.t().

    optional operation mask.

Return
  • mean: Evision.Mat.t().

    output parameter: calculated mean value.

  • stddev: Evision.Mat.t().

    output parameter: calculated standard deviation.

Calculates a mean and standard deviation of array elements. The function cv::meanStdDev calculates the mean and the standard deviation M of array elements independently for each channel and returns it via the output parameters: \f[\begin{array}{l} N = \sum _{I, \texttt{mask} (I) \ne 0} 1 \\ \texttt{mean} _c = \frac{\sum_{ I: \; \texttt{mask}(I) \ne 0} \texttt{src} (I)_c}{N} \\ \texttt{stddev} _c = \sqrt{\frac{\sum_{ I: \; \texttt{mask}(I) \ne 0} \left ( \texttt{src} (I)_c - \texttt{mean} _c \right )^2}{N}} \end{array}\f] When all the mask elements are 0's, the function returns mean=stddev=Scalar::all(0). Note: The calculated standard deviation is only the diagonal of the complete normalized covariance matrix. If the full matrix is needed, you can reshape the multi-channel array M x N to the single-channel array M*N x mtx.channels() (only possible when the matrix is continuous) and then pass the matrix to calcCovarMatrix . @sa countNonZero, mean, norm, minMaxLoc, calcCovarMatrix

Python prototype (for reference only):

meanStdDev(src[, mean[, stddev[, mask]]]) -> mean, stddev
@spec medianBlur(Evision.Mat.maybe_mat_in(), integer()) ::
  Evision.Mat.t() | {:error, String.t()}

Blurs an image using the median filter.

Positional Arguments
  • src: Evision.Mat.t().

    input 1-, 3-, or 4-channel image; when ksize is 3 or 5, the image depth should be CV_8U, CV_16U, or CV_32F, for larger aperture sizes, it can only be CV_8U.

  • ksize: int.

    aperture linear size; it must be odd and greater than 1, for example: 3, 5, 7 ...

Return
  • dst: Evision.Mat.t().

    destination array of the same size and type as src.

The function smoothes an image using the median filter with the \f$\texttt{ksize} \times \texttt{ksize}\f$ aperture. Each channel of a multi-channel image is processed independently. In-place operation is supported. Note: The median filter uses #BORDER_REPLICATE internally to cope with border pixels, see #BorderTypes @sa bilateralFilter, blur, boxFilter, GaussianBlur

Python prototype (for reference only):

medianBlur(src, ksize[, dst]) -> dst
Link to this function

medianBlur(src, ksize, opts)

View Source
@spec medianBlur(Evision.Mat.maybe_mat_in(), integer(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Blurs an image using the median filter.

Positional Arguments
  • src: Evision.Mat.t().

    input 1-, 3-, or 4-channel image; when ksize is 3 or 5, the image depth should be CV_8U, CV_16U, or CV_32F, for larger aperture sizes, it can only be CV_8U.

  • ksize: int.

    aperture linear size; it must be odd and greater than 1, for example: 3, 5, 7 ...

Return
  • dst: Evision.Mat.t().

    destination array of the same size and type as src.

The function smoothes an image using the median filter with the \f$\texttt{ksize} \times \texttt{ksize}\f$ aperture. Each channel of a multi-channel image is processed independently. In-place operation is supported. Note: The median filter uses #BORDER_REPLICATE internally to cope with border pixels, see #BorderTypes @sa bilateralFilter, blur, boxFilter, GaussianBlur

Python prototype (for reference only):

medianBlur(src, ksize[, dst]) -> dst
@spec merge([Evision.Mat.maybe_mat_in()]) :: Evision.Mat.t() | {:error, String.t()}

merge

Positional Arguments
  • mv: [Evision.Mat].

    input vector of matrices to be merged; all the matrices in mv must have the same size and the same depth.

Return
  • dst: Evision.Mat.t().

    output array of the same size and the same depth as mv[0]; The number of channels will be the total number of channels in the matrix array.

Has overloading in C++

Python prototype (for reference only):

merge(mv[, dst]) -> dst
@spec merge([Evision.Mat.maybe_mat_in()], [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

merge

Positional Arguments
  • mv: [Evision.Mat].

    input vector of matrices to be merged; all the matrices in mv must have the same size and the same depth.

Return
  • dst: Evision.Mat.t().

    output array of the same size and the same depth as mv[0]; The number of channels will be the total number of channels in the matrix array.

Has overloading in C++

Python prototype (for reference only):

merge(mv[, dst]) -> dst

Calculates per-element minimum of two arrays or an array and a scalar.

Positional Arguments
  • src1: Evision.Mat.t().

    first input array.

  • src2: Evision.Mat.t().

    second input array of the same size and type as src1.

Return
  • dst: Evision.Mat.t().

    output array of the same size and type as src1.

The function cv::min calculates the per-element minimum of two arrays: \f[\texttt{dst} (I)= \min ( \texttt{src1} (I), \texttt{src2} (I))\f] or array and a scalar: \f[\texttt{dst} (I)= \min ( \texttt{src1} (I), \texttt{value} )\f] @sa max, compare, inRange, minMaxLoc

Python prototype (for reference only):

min(src1, src2[, dst]) -> dst
@spec min(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Calculates per-element minimum of two arrays or an array and a scalar.

Positional Arguments
  • src1: Evision.Mat.t().

    first input array.

  • src2: Evision.Mat.t().

    second input array of the same size and type as src1.

Return
  • dst: Evision.Mat.t().

    output array of the same size and type as src1.

The function cv::min calculates the per-element minimum of two arrays: \f[\texttt{dst} (I)= \min ( \texttt{src1} (I), \texttt{src2} (I))\f] or array and a scalar: \f[\texttt{dst} (I)= \min ( \texttt{src1} (I), \texttt{value} )\f] @sa max, compare, inRange, minMaxLoc

Python prototype (for reference only):

min(src1, src2[, dst]) -> dst
@spec minAreaRect(Evision.Mat.maybe_mat_in()) ::
  Evision.RotatedRect.t() | {:error, String.t()}

Finds a rotated rectangle of the minimum area enclosing the input 2D point set.

Positional Arguments
  • points: Evision.Mat.t().

    Input vector of 2D points, stored in std::vector\<> or Mat

Return
  • retval: {centre={x, y}, size={s1, s2}, angle}

The function calculates and returns the minimum-area bounding rectangle (possibly rotated) for a specified point set. Developer should keep in mind that the returned RotatedRect can contain negative indices when data is close to the containing Mat element boundary.

Python prototype (for reference only):

minAreaRect(points) -> retval
Link to this function

minEnclosingCircle(points)

View Source
@spec minEnclosingCircle(Evision.Mat.maybe_mat_in()) ::
  {{number(), number()}, number()} | {:error, String.t()}

Finds a circle of the minimum area enclosing a 2D point set.

Positional Arguments
  • points: Evision.Mat.t().

    Input vector of 2D points, stored in std::vector\<> or Mat

Return
  • center: Point2f.

    Output center of the circle.

  • radius: float.

    Output radius of the circle.

The function finds the minimal enclosing circle of a 2D point set using an iterative algorithm.

Python prototype (for reference only):

minEnclosingCircle(points) -> center, radius
Link to this function

minEnclosingTriangle(points)

View Source
@spec minEnclosingTriangle(Evision.Mat.maybe_mat_in()) ::
  {number(), Evision.Mat.t()} | {:error, String.t()}

Finds a triangle of minimum area enclosing a 2D point set and returns its area.

Positional Arguments
  • points: Evision.Mat.t().

    Input vector of 2D points with depth CV_32S or CV_32F, stored in std::vector\<> or Mat

Return
  • retval: double

  • triangle: Evision.Mat.t().

    Output vector of three 2D points defining the vertices of the triangle. The depth of the OutputArray must be CV_32F.

The function finds a triangle of minimum area enclosing the given set of 2D points and returns its area. The output for a given 2D point set is shown in the image below. 2D points are depicted in red and the enclosing triangle in yellow*. Sample output of the minimum enclosing triangle function The implementation of the algorithm is based on O'Rourke's @cite ORourke86 and Klee and Laskowski's @cite KleeLaskowski85 papers. O'Rourke provides a \f$\theta(n)\f$ algorithm for finding the minimal enclosing triangle of a 2D convex polygon with n vertices. Since the #minEnclosingTriangle function takes a 2D point set as input an additional preprocessing step of computing the convex hull of the 2D point set is required. The complexity of the #convexHull function is \f$O(n log(n))\f$ which is higher than \f$\theta(n)\f$. Thus the overall complexity of the function is \f$O(n log(n))\f$.

Python prototype (for reference only):

minEnclosingTriangle(points[, triangle]) -> retval, triangle
Link to this function

minEnclosingTriangle(points, opts)

View Source
@spec minEnclosingTriangle(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  {number(), Evision.Mat.t()} | {:error, String.t()}

Finds a triangle of minimum area enclosing a 2D point set and returns its area.

Positional Arguments
  • points: Evision.Mat.t().

    Input vector of 2D points with depth CV_32S or CV_32F, stored in std::vector\<> or Mat

Return
  • retval: double

  • triangle: Evision.Mat.t().

    Output vector of three 2D points defining the vertices of the triangle. The depth of the OutputArray must be CV_32F.

The function finds a triangle of minimum area enclosing the given set of 2D points and returns its area. The output for a given 2D point set is shown in the image below. 2D points are depicted in red and the enclosing triangle in yellow*. Sample output of the minimum enclosing triangle function The implementation of the algorithm is based on O'Rourke's @cite ORourke86 and Klee and Laskowski's @cite KleeLaskowski85 papers. O'Rourke provides a \f$\theta(n)\f$ algorithm for finding the minimal enclosing triangle of a 2D convex polygon with n vertices. Since the #minEnclosingTriangle function takes a 2D point set as input an additional preprocessing step of computing the convex hull of the 2D point set is required. The complexity of the #convexHull function is \f$O(n log(n))\f$ which is higher than \f$\theta(n)\f$. Thus the overall complexity of the function is \f$O(n log(n))\f$.

Python prototype (for reference only):

minEnclosingTriangle(points[, triangle]) -> retval, triangle
@spec minMaxLoc(Evision.Mat.maybe_mat_in()) ::
  {number(), number(), {number(), number()}, {number(), number()}}
  | {:error, String.t()}

Finds the global minimum and maximum in an array.

Positional Arguments
  • src: Evision.Mat.t().

    input single-channel array.

Keyword Arguments
  • mask: Evision.Mat.t().

    optional mask used to select a sub-array.

Return
  • minVal: double*.

    pointer to the returned minimum value; NULL is used if not required.

  • maxVal: double*.

    pointer to the returned maximum value; NULL is used if not required.

  • minLoc: Point*.

    pointer to the returned minimum location (in 2D case); NULL is used if not required.

  • maxLoc: Point*.

    pointer to the returned maximum location (in 2D case); NULL is used if not required.

The function cv::minMaxLoc finds the minimum and maximum element values and their positions. The extremums are searched across the whole array or, if mask is not an empty array, in the specified array region. The function do not work with multi-channel arrays. If you need to find minimum or maximum elements across all the channels, use Mat::reshape first to reinterpret the array as single-channel. Or you may extract the particular channel using either extractImageCOI , or mixChannels , or split . @sa max, min, reduceArgMin, reduceArgMax, compare, inRange, extractImageCOI, mixChannels, split, Mat::reshape

Python prototype (for reference only):

minMaxLoc(src[, mask]) -> minVal, maxVal, minLoc, maxLoc
@spec minMaxLoc(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  {number(), number(), {number(), number()}, {number(), number()}}
  | {:error, String.t()}

Finds the global minimum and maximum in an array.

Positional Arguments
  • src: Evision.Mat.t().

    input single-channel array.

Keyword Arguments
  • mask: Evision.Mat.t().

    optional mask used to select a sub-array.

Return
  • minVal: double*.

    pointer to the returned minimum value; NULL is used if not required.

  • maxVal: double*.

    pointer to the returned maximum value; NULL is used if not required.

  • minLoc: Point*.

    pointer to the returned minimum location (in 2D case); NULL is used if not required.

  • maxLoc: Point*.

    pointer to the returned maximum location (in 2D case); NULL is used if not required.

The function cv::minMaxLoc finds the minimum and maximum element values and their positions. The extremums are searched across the whole array or, if mask is not an empty array, in the specified array region. The function do not work with multi-channel arrays. If you need to find minimum or maximum elements across all the channels, use Mat::reshape first to reinterpret the array as single-channel. Or you may extract the particular channel using either extractImageCOI , or mixChannels , or split . @sa max, min, reduceArgMin, reduceArgMax, compare, inRange, extractImageCOI, mixChannels, split, Mat::reshape

Python prototype (for reference only):

minMaxLoc(src[, mask]) -> minVal, maxVal, minLoc, maxLoc
Link to this function

mixChannels(src, dst, fromTo)

View Source
@spec mixChannels([Evision.Mat.maybe_mat_in()], [Evision.Mat.maybe_mat_in()], [
  integer()
]) ::
  [Evision.Mat.t()] | {:error, String.t()}

mixChannels

Positional Arguments
  • src: [Evision.Mat].

    input array or vector of matrices; all of the matrices must have the same size and the same depth.

  • fromTo: [int].

    array of index pairs specifying which channels are copied and where; fromTo[k*2] is a 0-based index of the input channel in src, fromTo[k*2+1] is an index of the output channel in dst; the continuous channel numbering is used: the first input image channels are indexed from 0 to src[0].channels()-1, the second input image channels are indexed from src[0].channels() to src[0].channels() + src[1].channels()-1, and so on, the same scheme is used for the output image channels; as a special case, when fromTo[k*2] is negative, the corresponding output channel is filled with zero .

Return
  • dst: [Evision.Mat].

    output array or vector of matrices; all the matrices must be allocated; their size and depth must be the same as in src[0].

Has overloading in C++

Python prototype (for reference only):

mixChannels(src, dst, fromTo) -> dst
@spec moments(Evision.Mat.maybe_mat_in()) :: map() | {:error, String.t()}

Calculates all of the moments up to the third order of a polygon or rasterized shape.

Positional Arguments
  • array: Evision.Mat.t().

    Raster image (single-channel, 8-bit or floating-point 2D array) or an array ( \f$1 \times N\f$ or \f$N \times 1\f$ ) of 2D points (Point or Point2f ).

Keyword Arguments
  • binaryImage: bool.

    If it is true, all non-zero image pixels are treated as 1's. The parameter is used for images only.

Return
  • retval: Moments

The function computes moments, up to the 3rd order, of a vector shape or a rasterized shape. The results are returned in the structure cv::Moments. @returns moments. Note: Only applicable to contour moments calculations from Python bindings: Note that the numpy type for the input array should be either np.int32 or np.float32. @sa contourArea, arcLength

Python prototype (for reference only):

moments(array[, binaryImage]) -> retval
@spec moments(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  map() | {:error, String.t()}

Calculates all of the moments up to the third order of a polygon or rasterized shape.

Positional Arguments
  • array: Evision.Mat.t().

    Raster image (single-channel, 8-bit or floating-point 2D array) or an array ( \f$1 \times N\f$ or \f$N \times 1\f$ ) of 2D points (Point or Point2f ).

Keyword Arguments
  • binaryImage: bool.

    If it is true, all non-zero image pixels are treated as 1's. The parameter is used for images only.

Return
  • retval: Moments

The function computes moments, up to the 3rd order, of a vector shape or a rasterized shape. The results are returned in the structure cv::Moments. @returns moments. Note: Only applicable to contour moments calculations from Python bindings: Note that the numpy type for the input array should be either np.int32 or np.float32. @sa contourArea, arcLength

Python prototype (for reference only):

moments(array[, binaryImage]) -> retval
Link to this function

morphologyEx(src, op, kernel)

View Source
@spec morphologyEx(Evision.Mat.maybe_mat_in(), integer(), Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

Performs advanced morphological transformations.

Positional Arguments
  • src: Evision.Mat.t().

    Source image. The number of channels can be arbitrary. The depth should be one of CV_8U, CV_16U, CV_16S, CV_32F or CV_64F.

  • op: int.

    Type of a morphological operation, see #MorphTypes

  • kernel: Evision.Mat.t().

    Structuring element. It can be created using #getStructuringElement.

Keyword Arguments
  • anchor: Point.

    Anchor position with the kernel. Negative values mean that the anchor is at the kernel center.

  • iterations: int.

    Number of times erosion and dilation are applied.

  • borderType: int.

    Pixel extrapolation method, see #BorderTypes. #BORDER_WRAP is not supported.

  • borderValue: Scalar.

    Border value in case of a constant border. The default value has a special meaning.

Return
  • dst: Evision.Mat.t().

    Destination image of the same size and type as source image.

The function cv::morphologyEx can perform advanced morphological transformations using an erosion and dilation as basic operations. Any of the operations can be done in-place. In case of multi-channel images, each channel is processed independently. @sa dilate, erode, getStructuringElement Note: The number of iterations is the number of times erosion or dilatation operation will be applied. For instance, an opening operation (#MORPH_OPEN) with two iterations is equivalent to apply successively: erode -> erode -> dilate -> dilate (and not erode -> dilate -> erode -> dilate).

Python prototype (for reference only):

morphologyEx(src, op, kernel[, dst[, anchor[, iterations[, borderType[, borderValue]]]]]) -> dst
Link to this function

morphologyEx(src, op, kernel, opts)

View Source
@spec morphologyEx(
  Evision.Mat.maybe_mat_in(),
  integer(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Performs advanced morphological transformations.

Positional Arguments
  • src: Evision.Mat.t().

    Source image. The number of channels can be arbitrary. The depth should be one of CV_8U, CV_16U, CV_16S, CV_32F or CV_64F.

  • op: int.

    Type of a morphological operation, see #MorphTypes

  • kernel: Evision.Mat.t().

    Structuring element. It can be created using #getStructuringElement.

Keyword Arguments
  • anchor: Point.

    Anchor position with the kernel. Negative values mean that the anchor is at the kernel center.

  • iterations: int.

    Number of times erosion and dilation are applied.

  • borderType: int.

    Pixel extrapolation method, see #BorderTypes. #BORDER_WRAP is not supported.

  • borderValue: Scalar.

    Border value in case of a constant border. The default value has a special meaning.

Return
  • dst: Evision.Mat.t().

    Destination image of the same size and type as source image.

The function cv::morphologyEx can perform advanced morphological transformations using an erosion and dilation as basic operations. Any of the operations can be done in-place. In case of multi-channel images, each channel is processed independently. @sa dilate, erode, getStructuringElement Note: The number of iterations is the number of times erosion or dilatation operation will be applied. For instance, an opening operation (#MORPH_OPEN) with two iterations is equivalent to apply successively: erode -> erode -> dilate -> dilate (and not erode -> dilate -> erode -> dilate).

Python prototype (for reference only):

morphologyEx(src, op, kernel[, dst[, anchor[, iterations[, borderType[, borderValue]]]]]) -> dst
Link to this function

moveWindow(winname, x, y)

View Source
@spec moveWindow(binary(), integer(), integer()) :: :ok | {:error, String.t()}

Moves the window to the specified position

Positional Arguments
  • winname: String.

    Name of the window.

  • x: int.

    The new x-coordinate of the window.

  • y: int.

    The new y-coordinate of the window.

Python prototype (for reference only):

moveWindow(winname, x, y) -> None
Link to this function

mulSpectrums(a, b, flags)

View Source
@spec mulSpectrums(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in(), integer()) ::
  Evision.Mat.t() | {:error, String.t()}

Performs the per-element multiplication of two Fourier spectrums.

Positional Arguments
  • a: Evision.Mat.t().

    first input array.

  • b: Evision.Mat.t().

    second input array of the same size and type as src1 .

  • flags: int.

    operation flags; currently, the only supported flag is cv::DFT_ROWS, which indicates that each row of src1 and src2 is an independent 1D Fourier spectrum. If you do not want to use this flag, then simply add a 0 as value.

Keyword Arguments
  • conjB: bool.

    optional flag that conjugates the second input array before the multiplication (true) or not (false).

Return
  • c: Evision.Mat.t().

    output array of the same size and type as src1 .

The function cv::mulSpectrums performs the per-element multiplication of the two CCS-packed or complex matrices that are results of a real or complex Fourier transform. The function, together with dft and idft , may be used to calculate convolution (pass conjB=false ) or correlation (pass conjB=true ) of two arrays rapidly. When the arrays are complex, they are simply multiplied (per element) with an optional conjugation of the second-array elements. When the arrays are real, they are assumed to be CCS-packed (see dft for details).

Python prototype (for reference only):

mulSpectrums(a, b, flags[, c[, conjB]]) -> c
Link to this function

mulSpectrums(a, b, flags, opts)

View Source
@spec mulSpectrums(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  integer(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Performs the per-element multiplication of two Fourier spectrums.

Positional Arguments
  • a: Evision.Mat.t().

    first input array.

  • b: Evision.Mat.t().

    second input array of the same size and type as src1 .

  • flags: int.

    operation flags; currently, the only supported flag is cv::DFT_ROWS, which indicates that each row of src1 and src2 is an independent 1D Fourier spectrum. If you do not want to use this flag, then simply add a 0 as value.

Keyword Arguments
  • conjB: bool.

    optional flag that conjugates the second input array before the multiplication (true) or not (false).

Return
  • c: Evision.Mat.t().

    output array of the same size and type as src1 .

The function cv::mulSpectrums performs the per-element multiplication of the two CCS-packed or complex matrices that are results of a real or complex Fourier transform. The function, together with dft and idft , may be used to calculate convolution (pass conjB=false ) or correlation (pass conjB=true ) of two arrays rapidly. When the arrays are complex, they are simply multiplied (per element) with an optional conjugation of the second-array elements. When the arrays are real, they are assumed to be CCS-packed (see dft for details).

Python prototype (for reference only):

mulSpectrums(a, b, flags[, c[, conjB]]) -> c
@spec mulTransposed(Evision.Mat.maybe_mat_in(), boolean()) ::
  Evision.Mat.t() | {:error, String.t()}

Calculates the product of a matrix and its transposition.

Positional Arguments
  • src: Evision.Mat.t().

    input single-channel matrix. Note that unlike gemm, the function can multiply not only floating-point matrices.

  • aTa: bool.

    Flag specifying the multiplication ordering. See the description below.

Keyword Arguments
  • delta: Evision.Mat.t().

    Optional delta matrix subtracted from src before the multiplication. When the matrix is empty ( delta=noArray() ), it is assumed to be zero, that is, nothing is subtracted. If it has the same size as src , it is simply subtracted. Otherwise, it is "repeated" (see repeat ) to cover the full src and then subtracted. Type of the delta matrix, when it is not empty, must be the same as the type of created output matrix. See the dtype parameter description below.

  • scale: double.

    Optional scale factor for the matrix product.

  • dtype: int.

    Optional type of the output matrix. When it is negative, the output matrix will have the same type as src . Otherwise, it will be type=CV_MAT_DEPTH(dtype) that should be either CV_32F or CV_64F .

Return
  • dst: Evision.Mat.t().

    output square matrix.

The function cv::mulTransposed calculates the product of src and its transposition: \f[\texttt{dst} = \texttt{scale} ( \texttt{src} - \texttt{delta} )^T ( \texttt{src} - \texttt{delta} )\f] if aTa=true , and \f[\texttt{dst} = \texttt{scale} ( \texttt{src} - \texttt{delta} ) ( \texttt{src} - \texttt{delta} )^T\f] otherwise. The function is used to calculate the covariance matrix. With zero delta, it can be used as a faster substitute for general matrix product A*B when B=A' @sa calcCovarMatrix, gemm, repeat, reduce

Python prototype (for reference only):

mulTransposed(src, aTa[, dst[, delta[, scale[, dtype]]]]) -> dst
Link to this function

mulTransposed(src, aTa, opts)

View Source
@spec mulTransposed(
  Evision.Mat.maybe_mat_in(),
  boolean(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Calculates the product of a matrix and its transposition.

Positional Arguments
  • src: Evision.Mat.t().

    input single-channel matrix. Note that unlike gemm, the function can multiply not only floating-point matrices.

  • aTa: bool.

    Flag specifying the multiplication ordering. See the description below.

Keyword Arguments
  • delta: Evision.Mat.t().

    Optional delta matrix subtracted from src before the multiplication. When the matrix is empty ( delta=noArray() ), it is assumed to be zero, that is, nothing is subtracted. If it has the same size as src , it is simply subtracted. Otherwise, it is "repeated" (see repeat ) to cover the full src and then subtracted. Type of the delta matrix, when it is not empty, must be the same as the type of created output matrix. See the dtype parameter description below.

  • scale: double.

    Optional scale factor for the matrix product.

  • dtype: int.

    Optional type of the output matrix. When it is negative, the output matrix will have the same type as src . Otherwise, it will be type=CV_MAT_DEPTH(dtype) that should be either CV_32F or CV_64F .

Return
  • dst: Evision.Mat.t().

    output square matrix.

The function cv::mulTransposed calculates the product of src and its transposition: \f[\texttt{dst} = \texttt{scale} ( \texttt{src} - \texttt{delta} )^T ( \texttt{src} - \texttt{delta} )\f] if aTa=true , and \f[\texttt{dst} = \texttt{scale} ( \texttt{src} - \texttt{delta} ) ( \texttt{src} - \texttt{delta} )^T\f] otherwise. The function is used to calculate the covariance matrix. With zero delta, it can be used as a faster substitute for general matrix product A*B when B=A' @sa calcCovarMatrix, gemm, repeat, reduce

Python prototype (for reference only):

mulTransposed(src, aTa[, dst[, delta[, scale[, dtype]]]]) -> dst

Calculates the per-element scaled product of two arrays.

Positional Arguments
  • src1: Evision.Mat.t().

    first input array.

  • src2: Evision.Mat.t().

    second input array of the same size and the same type as src1.

Keyword Arguments
  • scale: double.

    optional scale factor.

  • dtype: int.

    optional depth of the output array

Return
  • dst: Evision.Mat.t().

    output array of the same size and type as src1.

The function multiply calculates the per-element product of two arrays: \f[\texttt{dst} (I)= \texttt{saturate} ( \texttt{scale} \cdot \texttt{src1} (I) \cdot \texttt{src2} (I))\f] There is also a @ref MatrixExpressions -friendly variant of the first function. See Mat::mul . For a not-per-element matrix product, see gemm . Note: Saturation is not applied when the output array has the depth CV_32S. You may even get result of an incorrect sign in the case of overflow. Note: (Python) Be careful to difference behaviour between src1/src2 are single number and they are tuple/array. multiply(src,X) means multiply(src,(X,X,X,X)). multiply(src,(X,)) means multiply(src,(X,0,0,0)). @sa add, subtract, divide, scaleAdd, addWeighted, accumulate, accumulateProduct, accumulateSquare, Mat::convertTo

Python prototype (for reference only):

multiply(src1, src2[, dst[, scale[, dtype]]]) -> dst
Link to this function

multiply(src1, src2, opts)

View Source
@spec multiply(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Calculates the per-element scaled product of two arrays.

Positional Arguments
  • src1: Evision.Mat.t().

    first input array.

  • src2: Evision.Mat.t().

    second input array of the same size and the same type as src1.

Keyword Arguments
  • scale: double.

    optional scale factor.

  • dtype: int.

    optional depth of the output array

Return
  • dst: Evision.Mat.t().

    output array of the same size and type as src1.

The function multiply calculates the per-element product of two arrays: \f[\texttt{dst} (I)= \texttt{saturate} ( \texttt{scale} \cdot \texttt{src1} (I) \cdot \texttt{src2} (I))\f] There is also a @ref MatrixExpressions -friendly variant of the first function. See Mat::mul . For a not-per-element matrix product, see gemm . Note: Saturation is not applied when the output array has the depth CV_32S. You may even get result of an incorrect sign in the case of overflow. Note: (Python) Be careful to difference behaviour between src1/src2 are single number and they are tuple/array. multiply(src,X) means multiply(src,(X,X,X,X)). multiply(src,(X,)) means multiply(src,(X,0,0,0)). @sa add, subtract, divide, scaleAdd, addWeighted, accumulate, accumulateProduct, accumulateSquare, Mat::convertTo

Python prototype (for reference only):

multiply(src1, src2[, dst[, scale[, dtype]]]) -> dst
@spec namedWindow(binary()) :: :ok | {:error, String.t()}

Creates a window.

Positional Arguments
  • winname: String.

    Name of the window in the window caption that may be used as a window identifier.

Keyword Arguments
  • flags: int.

    Flags of the window. The supported flags are: (cv::WindowFlags)

The function namedWindow creates a window that can be used as a placeholder for images and trackbars. Created windows are referred to by their names. If a window with the same name already exists, the function does nothing. You can call cv::destroyWindow or cv::destroyAllWindows to close the window and de-allocate any associated memory usage. For a simple program, you do not really have to call these functions because all the resources and windows of the application are closed automatically by the operating system upon exit. Note: Qt backend supports additional flags:

  • WINDOW_NORMAL or WINDOW_AUTOSIZE: WINDOW_NORMAL enables you to resize the window, whereas WINDOW_AUTOSIZE adjusts automatically the window size to fit the displayed image (see imshow ), and you cannot change the window size manually.

  • WINDOW_FREERATIO or WINDOW_KEEPRATIO: WINDOW_FREERATIO adjusts the image with no respect to its ratio, whereas WINDOW_KEEPRATIO keeps the image ratio.

  • WINDOW_GUI_NORMAL or WINDOW_GUI_EXPANDED: WINDOW_GUI_NORMAL is the old way to draw the window without statusbar and toolbar, whereas WINDOW_GUI_EXPANDED is a new enhanced GUI. By default, flags == WINDOW_AUTOSIZE | WINDOW_KEEPRATIO | WINDOW_GUI_EXPANDED

Python prototype (for reference only):

namedWindow(winname[, flags]) -> None
Link to this function

namedWindow(winname, opts)

View Source
@spec namedWindow(binary(), [{atom(), term()}, ...] | nil) ::
  :ok | {:error, String.t()}

Creates a window.

Positional Arguments
  • winname: String.

    Name of the window in the window caption that may be used as a window identifier.

Keyword Arguments
  • flags: int.

    Flags of the window. The supported flags are: (cv::WindowFlags)

The function namedWindow creates a window that can be used as a placeholder for images and trackbars. Created windows are referred to by their names. If a window with the same name already exists, the function does nothing. You can call cv::destroyWindow or cv::destroyAllWindows to close the window and de-allocate any associated memory usage. For a simple program, you do not really have to call these functions because all the resources and windows of the application are closed automatically by the operating system upon exit. Note: Qt backend supports additional flags:

  • WINDOW_NORMAL or WINDOW_AUTOSIZE: WINDOW_NORMAL enables you to resize the window, whereas WINDOW_AUTOSIZE adjusts automatically the window size to fit the displayed image (see imshow ), and you cannot change the window size manually.

  • WINDOW_FREERATIO or WINDOW_KEEPRATIO: WINDOW_FREERATIO adjusts the image with no respect to its ratio, whereas WINDOW_KEEPRATIO keeps the image ratio.

  • WINDOW_GUI_NORMAL or WINDOW_GUI_EXPANDED: WINDOW_GUI_NORMAL is the old way to draw the window without statusbar and toolbar, whereas WINDOW_GUI_EXPANDED is a new enhanced GUI. By default, flags == WINDOW_AUTOSIZE | WINDOW_KEEPRATIO | WINDOW_GUI_EXPANDED

Python prototype (for reference only):

namedWindow(winname[, flags]) -> None
@spec norm(Evision.Mat.maybe_mat_in()) :: number() | {:error, String.t()}

Calculates the absolute norm of an array.

Positional Arguments
  • src1: Evision.Mat.t().

    first input array.

Keyword Arguments
  • normType: int.

    type of the norm (see #NormTypes).

  • mask: Evision.Mat.t().

    optional operation mask; it must have the same size as src1 and CV_8UC1 type.

Return
  • retval: double

This version of #norm calculates the absolute norm of src1. The type of norm to calculate is specified using #NormTypes. As example for one array consider the function \f$r(x)= \begin{pmatrix} x \\ 1-x \end{pmatrix}, x \in [-1;1]\f$. The \f$ L_{1}, L_{2} \f$ and \f$ L_{\infty} \f$ norm for the sample value \f$r(-1) = \begin{pmatrix} -1 \\ 2 \end{pmatrix}\f$ is calculated as follows \f{align} \| r(-1) \|{L_1} &= |-1| + |2| = 3 \\ \| r(-1) \|{L2} &= \sqrt{(-1)^{2} + (2)^{2}} = \sqrt{5} \\ \| r(-1) \|{L_\infty} &= \max(|-1|,|2|) = 2 \f} and for \f$r(0.5) = \begin{pmatrix} 0.5 \\ 0.5 \end{pmatrix}\f$ the calculation is \f{align} \| r(0.5) \|{L_1} &= |0.5| + |0.5| = 1 \\ \| r(0.5) \|{L2} &= \sqrt{(0.5)^{2} + (0.5)^{2}} = \sqrt{0.5} \\ \| r(0.5) \|{L\infty} &= \max(|0.5|,|0.5|) = 0.5. \f} The following graphic shows all values for the three norm functions \f$\| r(x) \|\{L_1}, \| r(x) \|_{L_2}\f$ and \f$\| r(x) \|_{L_\infty}\f$. It is notable that the \f$ L_{1} \f$ norm forms the upper and the \f$ L_{\infty} \f$ norm forms the lower border for the example function \f$ r(x) \f$. Graphs for the different norm functions from the above example When the mask parameter is specified and it is not empty, the norm is If normType is not specified, #NORM_L2 is used. calculated only over the region specified by the mask. Multi-channel input arrays are treated as single-channel arrays, that is, the results for all channels are combined. Hamming norms can only be calculated with CV_8U depth arrays.

Python prototype (for reference only):

norm(src1[, normType[, mask]]) -> retval
@spec norm(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  number() | {:error, String.t()}
@spec norm(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in()) ::
  number() | {:error, String.t()}

Variant 1:

Calculates an absolute difference norm or a relative difference norm.

Positional Arguments
  • src1: Evision.Mat.t().

    first input array.

  • src2: Evision.Mat.t().

    second input array of the same size and the same type as src1.

Keyword Arguments
  • normType: int.

    type of the norm (see #NormTypes).

  • mask: Evision.Mat.t().

    optional operation mask; it must have the same size as src1 and CV_8UC1 type.

Return
  • retval: double

This version of cv::norm calculates the absolute difference norm or the relative difference norm of arrays src1 and src2. The type of norm to calculate is specified using #NormTypes.

Python prototype (for reference only):

norm(src1, src2[, normType[, mask]]) -> retval

Variant 2:

Calculates the absolute norm of an array.

Positional Arguments
  • src1: Evision.Mat.t().

    first input array.

Keyword Arguments
  • normType: int.

    type of the norm (see #NormTypes).

  • mask: Evision.Mat.t().

    optional operation mask; it must have the same size as src1 and CV_8UC1 type.

Return
  • retval: double

This version of #norm calculates the absolute norm of src1. The type of norm to calculate is specified using #NormTypes. As example for one array consider the function \f$r(x)= \begin{pmatrix} x \\ 1-x \end{pmatrix}, x \in [-1;1]\f$. The \f$ L_{1}, L_{2} \f$ and \f$ L_{\infty} \f$ norm for the sample value \f$r(-1) = \begin{pmatrix} -1 \\ 2 \end{pmatrix}\f$ is calculated as follows \f{align} \| r(-1) \|{L_1} &= |-1| + |2| = 3 \\ \| r(-1) \|{L2} &= \sqrt{(-1)^{2} + (2)^{2}} = \sqrt{5} \\ \| r(-1) \|{L_\infty} &= \max(|-1|,|2|) = 2 \f} and for \f$r(0.5) = \begin{pmatrix} 0.5 \\ 0.5 \end{pmatrix}\f$ the calculation is \f{align} \| r(0.5) \|{L_1} &= |0.5| + |0.5| = 1 \\ \| r(0.5) \|{L2} &= \sqrt{(0.5)^{2} + (0.5)^{2}} = \sqrt{0.5} \\ \| r(0.5) \|{L\infty} &= \max(|0.5|,|0.5|) = 0.5. \f} The following graphic shows all values for the three norm functions \f$\| r(x) \|\{L_1}, \| r(x) \|_{L_2}\f$ and \f$\| r(x) \|_{L_\infty}\f$. It is notable that the \f$ L_{1} \f$ norm forms the upper and the \f$ L_{\infty} \f$ norm forms the lower border for the example function \f$ r(x) \f$. Graphs for the different norm functions from the above example When the mask parameter is specified and it is not empty, the norm is If normType is not specified, #NORM_L2 is used. calculated only over the region specified by the mask. Multi-channel input arrays are treated as single-channel arrays, that is, the results for all channels are combined. Hamming norms can only be calculated with CV_8U depth arrays.

Python prototype (for reference only):

norm(src1[, normType[, mask]]) -> retval
@spec norm(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  number() | {:error, String.t()}

Calculates an absolute difference norm or a relative difference norm.

Positional Arguments
  • src1: Evision.Mat.t().

    first input array.

  • src2: Evision.Mat.t().

    second input array of the same size and the same type as src1.

Keyword Arguments
  • normType: int.

    type of the norm (see #NormTypes).

  • mask: Evision.Mat.t().

    optional operation mask; it must have the same size as src1 and CV_8UC1 type.

Return
  • retval: double

This version of cv::norm calculates the absolute difference norm or the relative difference norm of arrays src1 and src2. The type of norm to calculate is specified using #NormTypes.

Python prototype (for reference only):

norm(src1, src2[, normType[, mask]]) -> retval
@spec normalize(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

Normalizes the norm or value range of an array.

Positional Arguments
  • src: Evision.Mat.t().

    input array.

Keyword Arguments
  • alpha: double.

    norm value to normalize to or the lower range boundary in case of the range normalization.

  • beta: double.

    upper range boundary in case of the range normalization; it is not used for the norm normalization.

  • norm_type: int.

    normalization type (see cv::NormTypes).

  • dtype: int.

    when negative, the output array has the same type as src; otherwise, it has the same number of channels as src and the depth =CV_MAT_DEPTH(dtype).

  • mask: Evision.Mat.t().

    optional operation mask.

Return
  • dst: Evision.Mat.t().

    output array of the same size as src .

The function cv::normalize normalizes scale and shift the input array elements so that \f[\| \texttt{dst} \| _{L_p}= \texttt{alpha}\f] (where p=Inf, 1 or 2) when normType=NORM_INF, NORM_L1, or NORM_L2, respectively; or so that \f[\min _I \texttt{dst} (I)= \texttt{alpha} , \, \, \max _I \texttt{dst} (I)= \texttt{beta}\f] when normType=NORM_MINMAX (for dense arrays only). The optional mask specifies a sub-array to be normalized. This means that the norm or min-n-max are calculated over the sub-array, and then this sub-array is modified to be normalized. If you want to only use the mask to calculate the norm or min-max but modify the whole array, you can use norm and Mat::convertTo. In case of sparse matrices, only the non-zero values are analyzed and transformed. Because of this, the range transformation for sparse matrices is not allowed since it can shift the zero level. Possible usage with some positive example data:

vector<double> positiveData = { 2.0, 8.0, 10.0 };
vector<double> normalizedData_l1, normalizedData_l2, normalizedData_inf, normalizedData_minmax;
// Norm to probability (total count)
// sum(numbers) = 20.0
// 2.0      0.1     (2.0/20.0)
// 8.0      0.4     (8.0/20.0)
// 10.0     0.5     (10.0/20.0)
normalize(positiveData, normalizedData_l1, 1.0, 0.0, NORM_L1);
// Norm to unit vector: ||positiveData|| = 1.0
// 2.0      0.15
// 8.0      0.62
// 10.0     0.77
normalize(positiveData, normalizedData_l2, 1.0, 0.0, NORM_L2);
// Norm to max element
// 2.0      0.2     (2.0/10.0)
// 8.0      0.8     (8.0/10.0)
// 10.0     1.0     (10.0/10.0)
normalize(positiveData, normalizedData_inf, 1.0, 0.0, NORM_INF);
// Norm to range [0.0;1.0]
// 2.0      0.0     (shift to left border)
// 8.0      0.75    (6.0/8.0)
// 10.0     1.0     (shift to right border)
normalize(positiveData, normalizedData_minmax, 1.0, 0.0, NORM_MINMAX);

@sa norm, Mat::convertTo, SparseMat::convertTo

Python prototype (for reference only):

normalize(src, dst[, alpha[, beta[, norm_type[, dtype[, mask]]]]]) -> dst
Link to this function

normalize(src, dst, opts)

View Source
@spec normalize(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Normalizes the norm or value range of an array.

Positional Arguments
  • src: Evision.Mat.t().

    input array.

Keyword Arguments
  • alpha: double.

    norm value to normalize to or the lower range boundary in case of the range normalization.

  • beta: double.

    upper range boundary in case of the range normalization; it is not used for the norm normalization.

  • norm_type: int.

    normalization type (see cv::NormTypes).

  • dtype: int.

    when negative, the output array has the same type as src; otherwise, it has the same number of channels as src and the depth =CV_MAT_DEPTH(dtype).

  • mask: Evision.Mat.t().

    optional operation mask.

Return
  • dst: Evision.Mat.t().

    output array of the same size as src .

The function cv::normalize normalizes scale and shift the input array elements so that \f[\| \texttt{dst} \| _{L_p}= \texttt{alpha}\f] (where p=Inf, 1 or 2) when normType=NORM_INF, NORM_L1, or NORM_L2, respectively; or so that \f[\min _I \texttt{dst} (I)= \texttt{alpha} , \, \, \max _I \texttt{dst} (I)= \texttt{beta}\f] when normType=NORM_MINMAX (for dense arrays only). The optional mask specifies a sub-array to be normalized. This means that the norm or min-n-max are calculated over the sub-array, and then this sub-array is modified to be normalized. If you want to only use the mask to calculate the norm or min-max but modify the whole array, you can use norm and Mat::convertTo. In case of sparse matrices, only the non-zero values are analyzed and transformed. Because of this, the range transformation for sparse matrices is not allowed since it can shift the zero level. Possible usage with some positive example data:

vector<double> positiveData = { 2.0, 8.0, 10.0 };
vector<double> normalizedData_l1, normalizedData_l2, normalizedData_inf, normalizedData_minmax;
// Norm to probability (total count)
// sum(numbers) = 20.0
// 2.0      0.1     (2.0/20.0)
// 8.0      0.4     (8.0/20.0)
// 10.0     0.5     (10.0/20.0)
normalize(positiveData, normalizedData_l1, 1.0, 0.0, NORM_L1);
// Norm to unit vector: ||positiveData|| = 1.0
// 2.0      0.15
// 8.0      0.62
// 10.0     0.77
normalize(positiveData, normalizedData_l2, 1.0, 0.0, NORM_L2);
// Norm to max element
// 2.0      0.2     (2.0/10.0)
// 8.0      0.8     (8.0/10.0)
// 10.0     1.0     (10.0/10.0)
normalize(positiveData, normalizedData_inf, 1.0, 0.0, NORM_INF);
// Norm to range [0.0;1.0]
// 2.0      0.0     (shift to left border)
// 8.0      0.75    (6.0/8.0)
// 10.0     1.0     (shift to right border)
normalize(positiveData, normalizedData_minmax, 1.0, 0.0, NORM_MINMAX);

@sa norm, Mat::convertTo, SparseMat::convertTo

Python prototype (for reference only):

normalize(src, dst[, alpha[, beta[, norm_type[, dtype[, mask]]]]]) -> dst
@spec patchNaNs(Evision.Mat.maybe_mat_in()) :: Evision.Mat.t() | {:error, String.t()}

Replaces NaNs by given number

Keyword Arguments
  • val: double.

    value to convert the NaNs

Return
  • a: Evision.Mat.t().

    input/output matrix (CV_32F type).

Python prototype (for reference only):

patchNaNs(a[, val]) -> a
@spec patchNaNs(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Replaces NaNs by given number

Keyword Arguments
  • val: double.

    value to convert the NaNs

Return
  • a: Evision.Mat.t().

    input/output matrix (CV_32F type).

Python prototype (for reference only):

patchNaNs(a[, val]) -> a
Link to this function

pcaBackProject(data, mean, eigenvectors)

View Source

PCABackProject

Positional Arguments
  • data: Evision.Mat.t()
  • mean: Evision.Mat.t()
  • eigenvectors: Evision.Mat.t()
Return
  • result: Evision.Mat.t().

wrap PCA::backProject

Python prototype (for reference only):

PCABackProject(data, mean, eigenvectors[, result]) -> result
Link to this function

pcaBackProject(data, mean, eigenvectors, opts)

View Source
@spec pcaBackProject(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

PCABackProject

Positional Arguments
  • data: Evision.Mat.t()
  • mean: Evision.Mat.t()
  • eigenvectors: Evision.Mat.t()
Return
  • result: Evision.Mat.t().

wrap PCA::backProject

Python prototype (for reference only):

PCABackProject(data, mean, eigenvectors[, result]) -> result

PCACompute2

Positional Arguments
  • data: Evision.Mat.t()
Keyword Arguments
  • maxComponents: int.
Return
  • mean: Evision.Mat.t()
  • eigenvectors: Evision.Mat.t().
  • eigenvalues: Evision.Mat.t().

wrap PCA::operator() and add eigenvalues output parameter

Python prototype (for reference only):

PCACompute2(data, mean[, eigenvectors[, eigenvalues[, maxComponents]]]) -> mean, eigenvectors, eigenvalues
Link to this function

pcaCompute2(data, mean, opts)

View Source
@spec pcaCompute2(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  {Evision.Mat.t(), Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}
@spec pcaCompute2(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in(), number()) ::
  {Evision.Mat.t(), Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Variant 1:

PCACompute2

Positional Arguments
  • data: Evision.Mat.t()
  • retainedVariance: double
Return
  • mean: Evision.Mat.t()
  • eigenvectors: Evision.Mat.t().
  • eigenvalues: Evision.Mat.t().

wrap PCA::operator() and add eigenvalues output parameter

Python prototype (for reference only):

PCACompute2(data, mean, retainedVariance[, eigenvectors[, eigenvalues]]) -> mean, eigenvectors, eigenvalues

Variant 2:

PCACompute2

Positional Arguments
  • data: Evision.Mat.t()
Keyword Arguments
  • maxComponents: int.
Return
  • mean: Evision.Mat.t()
  • eigenvectors: Evision.Mat.t().
  • eigenvalues: Evision.Mat.t().

wrap PCA::operator() and add eigenvalues output parameter

Python prototype (for reference only):

PCACompute2(data, mean[, eigenvectors[, eigenvalues[, maxComponents]]]) -> mean, eigenvectors, eigenvalues
Link to this function

pcaCompute2(data, mean, retainedVariance, opts)

View Source
@spec pcaCompute2(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  number(),
  [{atom(), term()}, ...] | nil
) :: {Evision.Mat.t(), Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

PCACompute2

Positional Arguments
  • data: Evision.Mat.t()
  • retainedVariance: double
Return
  • mean: Evision.Mat.t()
  • eigenvectors: Evision.Mat.t().
  • eigenvalues: Evision.Mat.t().

wrap PCA::operator() and add eigenvalues output parameter

Python prototype (for reference only):

PCACompute2(data, mean, retainedVariance[, eigenvectors[, eigenvalues]]) -> mean, eigenvectors, eigenvalues

PCACompute

Positional Arguments
  • data: Evision.Mat.t()
Keyword Arguments
  • maxComponents: int.
Return
  • mean: Evision.Mat.t()
  • eigenvectors: Evision.Mat.t().

wrap PCA::operator()

Python prototype (for reference only):

PCACompute(data, mean[, eigenvectors[, maxComponents]]) -> mean, eigenvectors
Link to this function

pcaCompute(data, mean, opts)

View Source
@spec pcaCompute(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}
@spec pcaCompute(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in(), number()) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Variant 1:

PCACompute

Positional Arguments
  • data: Evision.Mat.t()
  • retainedVariance: double
Return
  • mean: Evision.Mat.t()
  • eigenvectors: Evision.Mat.t().

wrap PCA::operator()

Python prototype (for reference only):

PCACompute(data, mean, retainedVariance[, eigenvectors]) -> mean, eigenvectors

Variant 2:

PCACompute

Positional Arguments
  • data: Evision.Mat.t()
Keyword Arguments
  • maxComponents: int.
Return
  • mean: Evision.Mat.t()
  • eigenvectors: Evision.Mat.t().

wrap PCA::operator()

Python prototype (for reference only):

PCACompute(data, mean[, eigenvectors[, maxComponents]]) -> mean, eigenvectors
Link to this function

pcaCompute(data, mean, retainedVariance, opts)

View Source
@spec pcaCompute(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  number(),
  [{atom(), term()}, ...] | nil
) :: {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

PCACompute

Positional Arguments
  • data: Evision.Mat.t()
  • retainedVariance: double
Return
  • mean: Evision.Mat.t()
  • eigenvectors: Evision.Mat.t().

wrap PCA::operator()

Python prototype (for reference only):

PCACompute(data, mean, retainedVariance[, eigenvectors]) -> mean, eigenvectors
Link to this function

pcaProject(data, mean, eigenvectors)

View Source

PCAProject

Positional Arguments
  • data: Evision.Mat.t()
  • mean: Evision.Mat.t()
  • eigenvectors: Evision.Mat.t()
Return
  • result: Evision.Mat.t().

wrap PCA::project

Python prototype (for reference only):

PCAProject(data, mean, eigenvectors[, result]) -> result
Link to this function

pcaProject(data, mean, eigenvectors, opts)

View Source
@spec pcaProject(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

PCAProject

Positional Arguments
  • data: Evision.Mat.t()
  • mean: Evision.Mat.t()
  • eigenvectors: Evision.Mat.t()
Return
  • result: Evision.Mat.t().

wrap PCA::project

Python prototype (for reference only):

PCAProject(data, mean, eigenvectors[, result]) -> result
@spec pencilSketch(Evision.Mat.maybe_mat_in()) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Pencil-like non-photorealistic line drawing

Positional Arguments
  • src: Evision.Mat.t().

    Input 8-bit 3-channel image.

Keyword Arguments
  • sigma_s: float.

    %Range between 0 to 200.

  • sigma_r: float.

    %Range between 0 to 1.

  • shade_factor: float.

    %Range between 0 to 0.1.

Return
  • dst1: Evision.Mat.t().

    Output 8-bit 1-channel image.

  • dst2: Evision.Mat.t().

    Output image with the same size and type as src.

Python prototype (for reference only):

pencilSketch(src[, dst1[, dst2[, sigma_s[, sigma_r[, shade_factor]]]]]) -> dst1, dst2
@spec pencilSketch(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Pencil-like non-photorealistic line drawing

Positional Arguments
  • src: Evision.Mat.t().

    Input 8-bit 3-channel image.

Keyword Arguments
  • sigma_s: float.

    %Range between 0 to 200.

  • sigma_r: float.

    %Range between 0 to 1.

  • shade_factor: float.

    %Range between 0 to 0.1.

Return
  • dst1: Evision.Mat.t().

    Output 8-bit 1-channel image.

  • dst2: Evision.Mat.t().

    Output image with the same size and type as src.

Python prototype (for reference only):

pencilSketch(src[, dst1[, dst2[, sigma_s[, sigma_r[, shade_factor]]]]]) -> dst1, dst2
Link to this function

perspectiveTransform(src, m)

View Source
@spec perspectiveTransform(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

Performs the perspective matrix transformation of vectors.

Positional Arguments
  • src: Evision.Mat.t().

    input two-channel or three-channel floating-point array; each element is a 2D/3D vector to be transformed.

  • m: Evision.Mat.t().

    3x3 or 4x4 floating-point transformation matrix.

Return
  • dst: Evision.Mat.t().

    output array of the same size and type as src.

The function cv::perspectiveTransform transforms every element of src by treating it as a 2D or 3D vector, in the following way: \f[(x, y, z) \rightarrow (x'/w, y'/w, z'/w)\f] where \f[(x', y', z', w') = \texttt{mat} \cdot \begin{bmatrix} x & y & z & 1 \end{bmatrix}\f] and \f[w = \fork{w'}{if (w' \ne 0)}{\infty}{otherwise}\f] Here a 3D vector transformation is shown. In case of a 2D vector transformation, the z component is omitted. Note: The function transforms a sparse set of 2D or 3D vectors. If you want to transform an image using perspective transformation, use warpPerspective . If you have an inverse problem, that is, you want to compute the most probable perspective transformation out of several pairs of corresponding points, you can use getPerspectiveTransform or findHomography . @sa transform, warpPerspective, getPerspectiveTransform, findHomography

Python prototype (for reference only):

perspectiveTransform(src, m[, dst]) -> dst
Link to this function

perspectiveTransform(src, m, opts)

View Source
@spec perspectiveTransform(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Performs the perspective matrix transformation of vectors.

Positional Arguments
  • src: Evision.Mat.t().

    input two-channel or three-channel floating-point array; each element is a 2D/3D vector to be transformed.

  • m: Evision.Mat.t().

    3x3 or 4x4 floating-point transformation matrix.

Return
  • dst: Evision.Mat.t().

    output array of the same size and type as src.

The function cv::perspectiveTransform transforms every element of src by treating it as a 2D or 3D vector, in the following way: \f[(x, y, z) \rightarrow (x'/w, y'/w, z'/w)\f] where \f[(x', y', z', w') = \texttt{mat} \cdot \begin{bmatrix} x & y & z & 1 \end{bmatrix}\f] and \f[w = \fork{w'}{if (w' \ne 0)}{\infty}{otherwise}\f] Here a 3D vector transformation is shown. In case of a 2D vector transformation, the z component is omitted. Note: The function transforms a sparse set of 2D or 3D vectors. If you want to transform an image using perspective transformation, use warpPerspective . If you have an inverse problem, that is, you want to compute the most probable perspective transformation out of several pairs of corresponding points, you can use getPerspectiveTransform or findHomography . @sa transform, warpPerspective, getPerspectiveTransform, findHomography

Python prototype (for reference only):

perspectiveTransform(src, m[, dst]) -> dst

Calculates the rotation angle of 2D vectors.

Positional Arguments
  • x: Evision.Mat.t().

    input floating-point array of x-coordinates of 2D vectors.

  • y: Evision.Mat.t().

    input array of y-coordinates of 2D vectors; it must have the same size and the same type as x.

Keyword Arguments
  • angleInDegrees: bool.

    when true, the function calculates the angle in degrees, otherwise, they are measured in radians.

Return
  • angle: Evision.Mat.t().

    output array of vector angles; it has the same size and same type as x .

The function cv::phase calculates the rotation angle of each 2D vector that is formed from the corresponding elements of x and y : \f[\texttt{angle} (I) = \texttt{atan2} ( \texttt{y} (I), \texttt{x} (I))\f] The angle estimation accuracy is about 0.3 degrees. When x(I)=y(I)=0 , the corresponding angle(I) is set to 0.

Python prototype (for reference only):

phase(x, y[, angle[, angleInDegrees]]) -> angle
@spec phase(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Calculates the rotation angle of 2D vectors.

Positional Arguments
  • x: Evision.Mat.t().

    input floating-point array of x-coordinates of 2D vectors.

  • y: Evision.Mat.t().

    input array of y-coordinates of 2D vectors; it must have the same size and the same type as x.

Keyword Arguments
  • angleInDegrees: bool.

    when true, the function calculates the angle in degrees, otherwise, they are measured in radians.

Return
  • angle: Evision.Mat.t().

    output array of vector angles; it has the same size and same type as x .

The function cv::phase calculates the rotation angle of each 2D vector that is formed from the corresponding elements of x and y : \f[\texttt{angle} (I) = \texttt{atan2} ( \texttt{y} (I), \texttt{x} (I))\f] The angle estimation accuracy is about 0.3 degrees. When x(I)=y(I)=0 , the corresponding angle(I) is set to 0.

Python prototype (for reference only):

phase(x, y[, angle[, angleInDegrees]]) -> angle
Link to this function

phaseCorrelate(src1, src2)

View Source
@spec phaseCorrelate(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in()) ::
  {{number(), number()}, number()} | {:error, String.t()}

The function is used to detect translational shifts that occur between two images.

Positional Arguments
  • src1: Evision.Mat.t().

    Source floating point array (CV_32FC1 or CV_64FC1)

  • src2: Evision.Mat.t().

    Source floating point array (CV_32FC1 or CV_64FC1)

Keyword Arguments
  • window: Evision.Mat.t().

    Floating point array with windowing coefficients to reduce edge effects (optional).

Return
  • retval: Point2d

  • response: double*.

    Signal power within the 5x5 centroid around the peak, between 0 and 1 (optional).

The operation takes advantage of the Fourier shift theorem for detecting the translational shift in the frequency domain. It can be used for fast image registration as well as motion estimation. For more information please see http://en.wikipedia.org/wiki/Phase_correlation Calculates the cross-power spectrum of two supplied source arrays. The arrays are padded if needed with getOptimalDFTSize. The function performs the following equations:

  • First it applies a Hanning window (see http://en.wikipedia.org/wiki/Hann_function) to each image to remove possible edge effects. This window is cached until the array size changes to speed up processing time.

  • Next it computes the forward DFTs of each source array: \f[\mathbf{G}_a = \mathcal{F}\{src_1\}, \; \mathbf{G}_b = \mathcal{F}\{src_2\}\f] where \f$\mathcal{F}\f$ is the forward DFT.

  • It then computes the cross-power spectrum of each frequency domain array: \f[R = \frac{ \mathbf{G}_a \mathbf{G}_b^*}{|\mathbf{G}_a \mathbf{G}_b^*|}\f]

  • Next the cross-correlation is converted back into the time domain via the inverse DFT: \f[r = \mathcal{F}^{-1}\{R\}\f]

  • Finally, it computes the peak location and computes a 5x5 weighted centroid around the peak to achieve sub-pixel accuracy. \f[(\Delta x, \Delta y) = \texttt{weightedCentroid} \{\arg \max_{(x, y)}\{r\}\}\f]

  • If non-zero, the response parameter is computed as the sum of the elements of r within the 5x5 centroid around the peak location. It is normalized to a maximum of 1 (meaning there is a single peak) and will be smaller when there are multiple peaks.

@returns detected phase shift (sub-pixel) between the two arrays. @sa dft, getOptimalDFTSize, idft, mulSpectrums createHanningWindow

Python prototype (for reference only):

phaseCorrelate(src1, src2[, window]) -> retval, response
Link to this function

phaseCorrelate(src1, src2, opts)

View Source
@spec phaseCorrelate(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: {{number(), number()}, number()} | {:error, String.t()}

The function is used to detect translational shifts that occur between two images.

Positional Arguments
  • src1: Evision.Mat.t().

    Source floating point array (CV_32FC1 or CV_64FC1)

  • src2: Evision.Mat.t().

    Source floating point array (CV_32FC1 or CV_64FC1)

Keyword Arguments
  • window: Evision.Mat.t().

    Floating point array with windowing coefficients to reduce edge effects (optional).

Return
  • retval: Point2d

  • response: double*.

    Signal power within the 5x5 centroid around the peak, between 0 and 1 (optional).

The operation takes advantage of the Fourier shift theorem for detecting the translational shift in the frequency domain. It can be used for fast image registration as well as motion estimation. For more information please see http://en.wikipedia.org/wiki/Phase_correlation Calculates the cross-power spectrum of two supplied source arrays. The arrays are padded if needed with getOptimalDFTSize. The function performs the following equations:

  • First it applies a Hanning window (see http://en.wikipedia.org/wiki/Hann_function) to each image to remove possible edge effects. This window is cached until the array size changes to speed up processing time.

  • Next it computes the forward DFTs of each source array: \f[\mathbf{G}_a = \mathcal{F}\{src_1\}, \; \mathbf{G}_b = \mathcal{F}\{src_2\}\f] where \f$\mathcal{F}\f$ is the forward DFT.

  • It then computes the cross-power spectrum of each frequency domain array: \f[R = \frac{ \mathbf{G}_a \mathbf{G}_b^*}{|\mathbf{G}_a \mathbf{G}_b^*|}\f]

  • Next the cross-correlation is converted back into the time domain via the inverse DFT: \f[r = \mathcal{F}^{-1}\{R\}\f]

  • Finally, it computes the peak location and computes a 5x5 weighted centroid around the peak to achieve sub-pixel accuracy. \f[(\Delta x, \Delta y) = \texttt{weightedCentroid} \{\arg \max_{(x, y)}\{r\}\}\f]

  • If non-zero, the response parameter is computed as the sum of the elements of r within the 5x5 centroid around the peak location. It is normalized to a maximum of 1 (meaning there is a single peak) and will be smaller when there are multiple peaks.

@returns detected phase shift (sub-pixel) between the two arrays. @sa dft, getOptimalDFTSize, idft, mulSpectrums createHanningWindow

Python prototype (for reference only):

phaseCorrelate(src1, src2[, window]) -> retval, response
Link to this function

pointPolygonTest(contour, pt, measureDist)

View Source
@spec pointPolygonTest(Evision.Mat.maybe_mat_in(), {number(), number()}, boolean()) ::
  number() | {:error, String.t()}

Performs a point-in-contour test.

Positional Arguments
  • contour: Evision.Mat.t().

    Input contour.

  • pt: Point2f.

    Point tested against the contour.

  • measureDist: bool.

    If true, the function estimates the signed distance from the point to the nearest contour edge. Otherwise, the function only checks if the point is inside a contour or not.

Return
  • retval: double

The function determines whether the point is inside a contour, outside, or lies on an edge (or coincides with a vertex). It returns positive (inside), negative (outside), or zero (on an edge) value, correspondingly. When measureDist=false , the return value is +1, -1, and 0, respectively. Otherwise, the return value is a signed distance between the point and the nearest contour edge. See below a sample output of the function where each image pixel is tested against the contour: sample output

Python prototype (for reference only):

pointPolygonTest(contour, pt, measureDist) -> retval
Link to this function

polarToCart(magnitude, angle)

View Source

Calculates x and y coordinates of 2D vectors from their magnitude and angle.

Positional Arguments
  • magnitude: Evision.Mat.t().

    input floating-point array of magnitudes of 2D vectors; it can be an empty matrix (=Mat()), in this case, the function assumes that all the magnitudes are =1; if it is not empty, it must have the same size and type as angle.

  • angle: Evision.Mat.t().

    input floating-point array of angles of 2D vectors.

Keyword Arguments
  • angleInDegrees: bool.

    when true, the input angles are measured in degrees, otherwise, they are measured in radians.

Return
  • x: Evision.Mat.t().

    output array of x-coordinates of 2D vectors; it has the same size and type as angle.

  • y: Evision.Mat.t().

    output array of y-coordinates of 2D vectors; it has the same size and type as angle.

The function cv::polarToCart calculates the Cartesian coordinates of each 2D vector represented by the corresponding elements of magnitude and angle: \f[\begin{array}{l} \texttt{x} (I) = \texttt{magnitude} (I) \cos ( \texttt{angle} (I)) \\ \texttt{y} (I) = \texttt{magnitude} (I) \sin ( \texttt{angle} (I)) \\ \end{array}\f] The relative accuracy of the estimated coordinates is about 1e-6. @sa cartToPolar, magnitude, phase, exp, log, pow, sqrt

Python prototype (for reference only):

polarToCart(magnitude, angle[, x[, y[, angleInDegrees]]]) -> x, y
Link to this function

polarToCart(magnitude, angle, opts)

View Source
@spec polarToCart(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Calculates x and y coordinates of 2D vectors from their magnitude and angle.

Positional Arguments
  • magnitude: Evision.Mat.t().

    input floating-point array of magnitudes of 2D vectors; it can be an empty matrix (=Mat()), in this case, the function assumes that all the magnitudes are =1; if it is not empty, it must have the same size and type as angle.

  • angle: Evision.Mat.t().

    input floating-point array of angles of 2D vectors.

Keyword Arguments
  • angleInDegrees: bool.

    when true, the input angles are measured in degrees, otherwise, they are measured in radians.

Return
  • x: Evision.Mat.t().

    output array of x-coordinates of 2D vectors; it has the same size and type as angle.

  • y: Evision.Mat.t().

    output array of y-coordinates of 2D vectors; it has the same size and type as angle.

The function cv::polarToCart calculates the Cartesian coordinates of each 2D vector represented by the corresponding elements of magnitude and angle: \f[\begin{array}{l} \texttt{x} (I) = \texttt{magnitude} (I) \cos ( \texttt{angle} (I)) \\ \texttt{y} (I) = \texttt{magnitude} (I) \sin ( \texttt{angle} (I)) \\ \end{array}\f] The relative accuracy of the estimated coordinates is about 1e-6. @sa cartToPolar, magnitude, phase, exp, log, pow, sqrt

Python prototype (for reference only):

polarToCart(magnitude, angle[, x[, y[, angleInDegrees]]]) -> x, y
@spec pollKey() :: integer() | {:error, String.t()}

Polls for a pressed key.

Return
  • retval: int

The function pollKey polls for a key event without waiting. It returns the code of the pressed key or -1 if no key was pressed since the last invocation. To wait until a key was pressed, use #waitKey. Note: The functions #waitKey and #pollKey are the only methods in HighGUI that can fetch and handle GUI events, so one of them needs to be called periodically for normal event processing unless HighGUI is used within an environment that takes care of event processing. Note: The function only works if there is at least one HighGUI window created and the window is active. If there are several HighGUI windows, any of them can be active.

Python prototype (for reference only):

pollKey() -> retval
Link to this function

polylines(img, pts, isClosed, color)

View Source
@spec polylines(
  Evision.Mat.maybe_mat_in(),
  [Evision.Mat.maybe_mat_in()],
  boolean(),
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()}
) :: Evision.Mat.t() | {:error, String.t()}

Draws several polygonal curves.

Positional Arguments
  • pts: [Evision.Mat].

    Array of polygonal curves.

  • isClosed: bool.

    Flag indicating whether the drawn polylines are closed or not. If they are closed, the function draws a line from the last vertex of each curve to its first vertex.

  • color: Scalar.

    Polyline color.

Keyword Arguments
  • thickness: int.

    Thickness of the polyline edges.

  • lineType: int.

    Type of the line segments. See #LineTypes

  • shift: int.

    Number of fractional bits in the vertex coordinates.

Return
  • img: Evision.Mat.t().

    Image.

The function cv::polylines draws one or more polygonal curves.

Python prototype (for reference only):

polylines(img, pts, isClosed, color[, thickness[, lineType[, shift]]]) -> img
Link to this function

polylines(img, pts, isClosed, color, opts)

View Source
@spec polylines(
  Evision.Mat.maybe_mat_in(),
  [Evision.Mat.maybe_mat_in()],
  boolean(),
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()},
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Draws several polygonal curves.

Positional Arguments
  • pts: [Evision.Mat].

    Array of polygonal curves.

  • isClosed: bool.

    Flag indicating whether the drawn polylines are closed or not. If they are closed, the function draws a line from the last vertex of each curve to its first vertex.

  • color: Scalar.

    Polyline color.

Keyword Arguments
  • thickness: int.

    Thickness of the polyline edges.

  • lineType: int.

    Type of the line segments. See #LineTypes

  • shift: int.

    Number of fractional bits in the vertex coordinates.

Return
  • img: Evision.Mat.t().

    Image.

The function cv::polylines draws one or more polygonal curves.

Python prototype (for reference only):

polylines(img, pts, isClosed, color[, thickness[, lineType[, shift]]]) -> img
@spec pow(Evision.Mat.maybe_mat_in(), number()) ::
  Evision.Mat.t() | {:error, String.t()}

Raises every array element to a power.

Positional Arguments
  • src: Evision.Mat.t().

    input array.

  • power: double.

    exponent of power.

Return
  • dst: Evision.Mat.t().

    output array of the same size and type as src.

The function cv::pow raises every element of the input array to power : \f[\texttt{dst} (I) = \fork{\texttt{src}(I)^{power}}{if (\texttt{power}) is integer}{|\texttt{src}(I)|^{power}}{otherwise}\f] So, for a non-integer power exponent, the absolute values of input array elements are used. However, it is possible to get true values for negative values using some extra operations. In the example below, computing the 5th root of array src shows:

Mat mask = src < 0;
pow(src, 1./5, dst);
subtract(Scalar::all(0), dst, dst, mask);

For some values of power, such as integer values, 0.5 and -0.5, specialized faster algorithms are used. Special values (NaN, Inf) are not handled. @sa sqrt, exp, log, cartToPolar, polarToCart

Python prototype (for reference only):

pow(src, power[, dst]) -> dst
@spec pow(Evision.Mat.maybe_mat_in(), number(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Raises every array element to a power.

Positional Arguments
  • src: Evision.Mat.t().

    input array.

  • power: double.

    exponent of power.

Return
  • dst: Evision.Mat.t().

    output array of the same size and type as src.

The function cv::pow raises every element of the input array to power : \f[\texttt{dst} (I) = \fork{\texttt{src}(I)^{power}}{if (\texttt{power}) is integer}{|\texttt{src}(I)|^{power}}{otherwise}\f] So, for a non-integer power exponent, the absolute values of input array elements are used. However, it is possible to get true values for negative values using some extra operations. In the example below, computing the 5th root of array src shows:

Mat mask = src < 0;
pow(src, 1./5, dst);
subtract(Scalar::all(0), dst, dst, mask);

For some values of power, such as integer values, 0.5 and -0.5, specialized faster algorithms are used. Special values (NaN, Inf) are not handled. @sa sqrt, exp, log, cartToPolar, polarToCart

Python prototype (for reference only):

pow(src, power[, dst]) -> dst
Link to this function

preCornerDetect(src, ksize)

View Source
@spec preCornerDetect(Evision.Mat.maybe_mat_in(), integer()) ::
  Evision.Mat.t() | {:error, String.t()}

Calculates a feature map for corner detection.

Positional Arguments
  • src: Evision.Mat.t().

    Source single-channel 8-bit of floating-point image.

  • ksize: int.

    %Aperture size of the Sobel .

Keyword Arguments
  • borderType: int.

    Pixel extrapolation method. See #BorderTypes. #BORDER_WRAP is not supported.

Return
  • dst: Evision.Mat.t().

    Output image that has the type CV_32F and the same size as src .

The function calculates the complex spatial derivative-based function of the source image \f[\texttt{dst} = (D_x \texttt{src} )^2 \cdot D_{yy} \texttt{src} + (D_y \texttt{src} )^2 \cdot D_{xx} \texttt{src} - 2 D_x \texttt{src} \cdot D_y \texttt{src} \cdot D_{xy} \texttt{src}\f] where \f$D_x\f$,\f$D_y\f$ are the first image derivatives, \f$D_{xx}\f$,\f$D_{yy}\f$ are the second image derivatives, and \f$D_{xy}\f$ is the mixed derivative. The corners can be found as local maximums of the functions, as shown below:

Mat corners, dilated_corners;
preCornerDetect(image, corners, 3);
// dilation with 3x3 rectangular structuring element
dilate(corners, dilated_corners, Mat(), 1);
Mat corner_mask = corners == dilated_corners;

Python prototype (for reference only):

preCornerDetect(src, ksize[, dst[, borderType]]) -> dst
Link to this function

preCornerDetect(src, ksize, opts)

View Source
@spec preCornerDetect(
  Evision.Mat.maybe_mat_in(),
  integer(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Calculates a feature map for corner detection.

Positional Arguments
  • src: Evision.Mat.t().

    Source single-channel 8-bit of floating-point image.

  • ksize: int.

    %Aperture size of the Sobel .

Keyword Arguments
  • borderType: int.

    Pixel extrapolation method. See #BorderTypes. #BORDER_WRAP is not supported.

Return
  • dst: Evision.Mat.t().

    Output image that has the type CV_32F and the same size as src .

The function calculates the complex spatial derivative-based function of the source image \f[\texttt{dst} = (D_x \texttt{src} )^2 \cdot D_{yy} \texttt{src} + (D_y \texttt{src} )^2 \cdot D_{xx} \texttt{src} - 2 D_x \texttt{src} \cdot D_y \texttt{src} \cdot D_{xy} \texttt{src}\f] where \f$D_x\f$,\f$D_y\f$ are the first image derivatives, \f$D_{xx}\f$,\f$D_{yy}\f$ are the second image derivatives, and \f$D_{xy}\f$ is the mixed derivative. The corners can be found as local maximums of the functions, as shown below:

Mat corners, dilated_corners;
preCornerDetect(image, corners, 3);
// dilation with 3x3 rectangular structuring element
dilate(corners, dilated_corners, Mat(), 1);
Mat corner_mask = corners == dilated_corners;

Python prototype (for reference only):

preCornerDetect(src, ksize[, dst[, borderType]]) -> dst
Link to this function

projectPoints(objectPoints, rvec, tvec, cameraMatrix, distCoeffs)

View Source

Projects 3D points to an image plane.

Positional Arguments
  • objectPoints: Evision.Mat.t().

    Array of object points expressed wrt. the world coordinate frame. A 3xN/Nx3 1-channel or 1xN/Nx1 3-channel (or vector\<Point3f> ), where N is the number of points in the view.

  • rvec: Evision.Mat.t().

    The rotation vector (@ref Rodrigues) that, together with tvec, performs a change of basis from world to camera coordinate system, see @ref calibrateCamera for details.

  • tvec: Evision.Mat.t().

    The translation vector, see parameter description above.

  • cameraMatrix: Evision.Mat.t().

    Camera intrinsic matrix \f$\cameramatrix{A}\f$ .

  • distCoeffs: Evision.Mat.t().

    Input vector of distortion coefficients \f$\distcoeffs\f$ . If the vector is empty, the zero distortion coefficients are assumed.

Keyword Arguments
  • aspectRatio: double.

    Optional "fixed aspect ratio" parameter. If the parameter is not 0, the function assumes that the aspect ratio (\f$f_x / f_y\f$) is fixed and correspondingly adjusts the jacobian matrix.

Return
  • imagePoints: Evision.Mat.t().

    Output array of image points, 1xN/Nx1 2-channel, or vector\<Point2f> .

  • jacobian: Evision.Mat.t().

    Optional output 2Nx(10+\<numDistCoeffs>) jacobian matrix of derivatives of image points with respect to components of the rotation vector, translation vector, focal lengths, coordinates of the principal point and the distortion coefficients. In the old interface different components of the jacobian are returned via different output parameters.

The function computes the 2D projections of 3D points to the image plane, given intrinsic and extrinsic camera parameters. Optionally, the function computes Jacobians -matrices of partial derivatives of image points coordinates (as functions of all the input parameters) with respect to the particular parameters, intrinsic and/or extrinsic. The Jacobians are used during the global optimization in @ref calibrateCamera, @ref solvePnP, and @ref stereoCalibrate. The function itself can also be used to compute a re-projection error, given the current intrinsic and extrinsic parameters. Note: By setting rvec = tvec = \f$[0, 0, 0]\f$, or by setting cameraMatrix to a 3x3 identity matrix, or by passing zero distortion coefficients, one can get various useful partial cases of the function. This means, one can compute the distorted coordinates for a sparse set of points or apply a perspective transformation (and also compute the derivatives) in the ideal zero-distortion setup.

Python prototype (for reference only):

projectPoints(objectPoints, rvec, tvec, cameraMatrix, distCoeffs[, imagePoints[, jacobian[, aspectRatio]]]) -> imagePoints, jacobian
Link to this function

projectPoints(objectPoints, rvec, tvec, cameraMatrix, distCoeffs, opts)

View Source

Projects 3D points to an image plane.

Positional Arguments
  • objectPoints: Evision.Mat.t().

    Array of object points expressed wrt. the world coordinate frame. A 3xN/Nx3 1-channel or 1xN/Nx1 3-channel (or vector\<Point3f> ), where N is the number of points in the view.

  • rvec: Evision.Mat.t().

    The rotation vector (@ref Rodrigues) that, together with tvec, performs a change of basis from world to camera coordinate system, see @ref calibrateCamera for details.

  • tvec: Evision.Mat.t().

    The translation vector, see parameter description above.

  • cameraMatrix: Evision.Mat.t().

    Camera intrinsic matrix \f$\cameramatrix{A}\f$ .

  • distCoeffs: Evision.Mat.t().

    Input vector of distortion coefficients \f$\distcoeffs\f$ . If the vector is empty, the zero distortion coefficients are assumed.

Keyword Arguments
  • aspectRatio: double.

    Optional "fixed aspect ratio" parameter. If the parameter is not 0, the function assumes that the aspect ratio (\f$f_x / f_y\f$) is fixed and correspondingly adjusts the jacobian matrix.

Return
  • imagePoints: Evision.Mat.t().

    Output array of image points, 1xN/Nx1 2-channel, or vector\<Point2f> .

  • jacobian: Evision.Mat.t().

    Optional output 2Nx(10+\<numDistCoeffs>) jacobian matrix of derivatives of image points with respect to components of the rotation vector, translation vector, focal lengths, coordinates of the principal point and the distortion coefficients. In the old interface different components of the jacobian are returned via different output parameters.

The function computes the 2D projections of 3D points to the image plane, given intrinsic and extrinsic camera parameters. Optionally, the function computes Jacobians -matrices of partial derivatives of image points coordinates (as functions of all the input parameters) with respect to the particular parameters, intrinsic and/or extrinsic. The Jacobians are used during the global optimization in @ref calibrateCamera, @ref solvePnP, and @ref stereoCalibrate. The function itself can also be used to compute a re-projection error, given the current intrinsic and extrinsic parameters. Note: By setting rvec = tvec = \f$[0, 0, 0]\f$, or by setting cameraMatrix to a 3x3 identity matrix, or by passing zero distortion coefficients, one can get various useful partial cases of the function. This means, one can compute the distorted coordinates for a sparse set of points or apply a perspective transformation (and also compute the derivatives) in the ideal zero-distortion setup.

Python prototype (for reference only):

projectPoints(objectPoints, rvec, tvec, cameraMatrix, distCoeffs[, imagePoints[, jacobian[, aspectRatio]]]) -> imagePoints, jacobian

Computes the Peak Signal-to-Noise Ratio (PSNR) image quality metric.

Positional Arguments
  • src1: Evision.Mat.t().

    first input array.

  • src2: Evision.Mat.t().

    second input array of the same size as src1.

Keyword Arguments
  • r: double.

    the maximum pixel value (255 by default)

Return
  • retval: double

This function calculates the Peak Signal-to-Noise Ratio (PSNR) image quality metric in decibels (dB), between two input arrays src1 and src2. The arrays must have the same type. The PSNR is calculated as follows: \f[ \texttt{PSNR} = 10 \cdot \log_{10}{\left( \frac{R^2}{MSE} \right) } \f] where R is the maximum integer value of depth (e.g. 255 in the case of CV_8U data) and MSE is the mean squared error between the two arrays.

Python prototype (for reference only):

PSNR(src1, src2[, R]) -> retval
@spec psnr(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  number() | {:error, String.t()}

Computes the Peak Signal-to-Noise Ratio (PSNR) image quality metric.

Positional Arguments
  • src1: Evision.Mat.t().

    first input array.

  • src2: Evision.Mat.t().

    second input array of the same size as src1.

Keyword Arguments
  • r: double.

    the maximum pixel value (255 by default)

Return
  • retval: double

This function calculates the Peak Signal-to-Noise Ratio (PSNR) image quality metric in decibels (dB), between two input arrays src1 and src2. The arrays must have the same type. The PSNR is calculated as follows: \f[ \texttt{PSNR} = 10 \cdot \log_{10}{\left( \frac{R^2}{MSE} \right) } \f] where R is the maximum integer value of depth (e.g. 255 in the case of CV_8U data) and MSE is the mean squared error between the two arrays.

Python prototype (for reference only):

PSNR(src1, src2[, R]) -> retval
Link to this function

putText(img, text, org, fontFace, fontScale, color)

View Source
@spec putText(
  Evision.Mat.maybe_mat_in(),
  binary(),
  {number(), number()},
  integer(),
  number(),
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()}
) :: Evision.Mat.t() | {:error, String.t()}

Draws a text string.

Positional Arguments
  • text: String.

    Text string to be drawn.

  • org: Point.

    Bottom-left corner of the text string in the image.

  • fontFace: int.

    Font type, see #HersheyFonts.

  • fontScale: double.

    Font scale factor that is multiplied by the font-specific base size.

  • color: Scalar.

    Text color.

Keyword Arguments
  • thickness: int.

    Thickness of the lines used to draw a text.

  • lineType: int.

    Line type. See #LineTypes

  • bottomLeftOrigin: bool.

    When true, the image data origin is at the bottom-left corner. Otherwise, it is at the top-left corner.

Return
  • img: Evision.Mat.t().

    Image.

The function cv::putText renders the specified text string in the image. Symbols that cannot be rendered using the specified font are replaced by question marks. See #getTextSize for a text rendering code example.

Python prototype (for reference only):

putText(img, text, org, fontFace, fontScale, color[, thickness[, lineType[, bottomLeftOrigin]]]) -> img
Link to this function

putText(img, text, org, fontFace, fontScale, color, opts)

View Source
@spec putText(
  Evision.Mat.maybe_mat_in(),
  binary(),
  {number(), number()},
  integer(),
  number(),
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()},
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Draws a text string.

Positional Arguments
  • text: String.

    Text string to be drawn.

  • org: Point.

    Bottom-left corner of the text string in the image.

  • fontFace: int.

    Font type, see #HersheyFonts.

  • fontScale: double.

    Font scale factor that is multiplied by the font-specific base size.

  • color: Scalar.

    Text color.

Keyword Arguments
  • thickness: int.

    Thickness of the lines used to draw a text.

  • lineType: int.

    Line type. See #LineTypes

  • bottomLeftOrigin: bool.

    When true, the image data origin is at the bottom-left corner. Otherwise, it is at the top-left corner.

Return
  • img: Evision.Mat.t().

    Image.

The function cv::putText renders the specified text string in the image. Symbols that cannot be rendered using the specified font are replaced by question marks. See #getTextSize for a text rendering code example.

Python prototype (for reference only):

putText(img, text, org, fontFace, fontScale, color[, thickness[, lineType[, bottomLeftOrigin]]]) -> img
@spec pyrDown(Evision.Mat.maybe_mat_in()) :: Evision.Mat.t() | {:error, String.t()}

Blurs an image and downsamples it.

Positional Arguments
  • src: Evision.Mat.t().

    input image.

Keyword Arguments
  • dstsize: Size.

    size of the output image.

  • borderType: int.

    Pixel extrapolation method, see #BorderTypes (#BORDER_CONSTANT isn't supported)

Return
  • dst: Evision.Mat.t().

    output image; it has the specified size and the same type as src.

By default, size of the output image is computed as Size((src.cols+1)/2, (src.rows+1)/2), but in any case, the following conditions should be satisfied: \f[\begin{array}{l} | \texttt{dstsize.width} *2-src.cols| \leq 2 \\ | \texttt{dstsize.height} *2-src.rows| \leq 2 \end{array}\f] The function performs the downsampling step of the Gaussian pyramid construction. First, it convolves the source image with the kernel: \f[\frac{1}{256} \begin{bmatrix} 1 & 4 & 6 & 4 & 1 \\ 4 & 16 & 24 & 16 & 4 \\ 6 & 24 & 36 & 24 & 6 \\ 4 & 16 & 24 & 16 & 4 \\ 1 & 4 & 6 & 4 & 1 \end{bmatrix}\f] Then, it downsamples the image by rejecting even rows and columns.

Python prototype (for reference only):

pyrDown(src[, dst[, dstsize[, borderType]]]) -> dst
@spec pyrDown(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Blurs an image and downsamples it.

Positional Arguments
  • src: Evision.Mat.t().

    input image.

Keyword Arguments
  • dstsize: Size.

    size of the output image.

  • borderType: int.

    Pixel extrapolation method, see #BorderTypes (#BORDER_CONSTANT isn't supported)

Return
  • dst: Evision.Mat.t().

    output image; it has the specified size and the same type as src.

By default, size of the output image is computed as Size((src.cols+1)/2, (src.rows+1)/2), but in any case, the following conditions should be satisfied: \f[\begin{array}{l} | \texttt{dstsize.width} *2-src.cols| \leq 2 \\ | \texttt{dstsize.height} *2-src.rows| \leq 2 \end{array}\f] The function performs the downsampling step of the Gaussian pyramid construction. First, it convolves the source image with the kernel: \f[\frac{1}{256} \begin{bmatrix} 1 & 4 & 6 & 4 & 1 \\ 4 & 16 & 24 & 16 & 4 \\ 6 & 24 & 36 & 24 & 6 \\ 4 & 16 & 24 & 16 & 4 \\ 1 & 4 & 6 & 4 & 1 \end{bmatrix}\f] Then, it downsamples the image by rejecting even rows and columns.

Python prototype (for reference only):

pyrDown(src[, dst[, dstsize[, borderType]]]) -> dst
Link to this function

pyrMeanShiftFiltering(src, sp, sr)

View Source
@spec pyrMeanShiftFiltering(Evision.Mat.maybe_mat_in(), number(), number()) ::
  Evision.Mat.t() | {:error, String.t()}

Performs initial step of meanshift segmentation of an image.

Positional Arguments
  • src: Evision.Mat.t().

    The source 8-bit, 3-channel image.

  • sp: double.

    The spatial window radius.

  • sr: double.

    The color window radius.

Keyword Arguments
  • maxLevel: int.

    Maximum level of the pyramid for the segmentation.

  • termcrit: TermCriteria.

    Termination criteria: when to stop meanshift iterations.

Return
  • dst: Evision.Mat.t().

    The destination image of the same format and the same size as the source.

The function implements the filtering stage of meanshift segmentation, that is, the output of the function is the filtered "posterized" image with color gradients and fine-grain texture flattened. At every pixel (X,Y) of the input image (or down-sized input image, see below) the function executes meanshift iterations, that is, the pixel (X,Y) neighborhood in the joint space-color hyperspace is considered: \f[(x,y): X- \texttt{sp} \le x \le X+ \texttt{sp} , Y- \texttt{sp} \le y \le Y+ \texttt{sp} , ||(R,G,B)-(r,g,b)|| \le \texttt{sr}\f] where (R,G,B) and (r,g,b) are the vectors of color components at (X,Y) and (x,y), respectively (though, the algorithm does not depend on the color space used, so any 3-component color space can be used instead). Over the neighborhood the average spatial value (X',Y') and average color vector (R',G',B') are found and they act as the neighborhood center on the next iteration: \f[(X,Y)~(X',Y'), (R,G,B)~(R',G',B').\f] After the iterations over, the color components of the initial pixel (that is, the pixel from where the iterations started) are set to the final value (average color at the last iteration): \f[I(X,Y) <- (R*,G*,B*)\f] When maxLevel > 0, the gaussian pyramid of maxLevel+1 levels is built, and the above procedure is run on the smallest layer first. After that, the results are propagated to the larger layer and the iterations are run again only on those pixels where the layer colors differ by more than sr from the lower-resolution layer of the pyramid. That makes boundaries of color regions sharper. Note that the results will be actually different from the ones obtained by running the meanshift procedure on the whole original image (i.e. when maxLevel==0).

Python prototype (for reference only):

pyrMeanShiftFiltering(src, sp, sr[, dst[, maxLevel[, termcrit]]]) -> dst
Link to this function

pyrMeanShiftFiltering(src, sp, sr, opts)

View Source
@spec pyrMeanShiftFiltering(
  Evision.Mat.maybe_mat_in(),
  number(),
  number(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Performs initial step of meanshift segmentation of an image.

Positional Arguments
  • src: Evision.Mat.t().

    The source 8-bit, 3-channel image.

  • sp: double.

    The spatial window radius.

  • sr: double.

    The color window radius.

Keyword Arguments
  • maxLevel: int.

    Maximum level of the pyramid for the segmentation.

  • termcrit: TermCriteria.

    Termination criteria: when to stop meanshift iterations.

Return
  • dst: Evision.Mat.t().

    The destination image of the same format and the same size as the source.

The function implements the filtering stage of meanshift segmentation, that is, the output of the function is the filtered "posterized" image with color gradients and fine-grain texture flattened. At every pixel (X,Y) of the input image (or down-sized input image, see below) the function executes meanshift iterations, that is, the pixel (X,Y) neighborhood in the joint space-color hyperspace is considered: \f[(x,y): X- \texttt{sp} \le x \le X+ \texttt{sp} , Y- \texttt{sp} \le y \le Y+ \texttt{sp} , ||(R,G,B)-(r,g,b)|| \le \texttt{sr}\f] where (R,G,B) and (r,g,b) are the vectors of color components at (X,Y) and (x,y), respectively (though, the algorithm does not depend on the color space used, so any 3-component color space can be used instead). Over the neighborhood the average spatial value (X',Y') and average color vector (R',G',B') are found and they act as the neighborhood center on the next iteration: \f[(X,Y)~(X',Y'), (R,G,B)~(R',G',B').\f] After the iterations over, the color components of the initial pixel (that is, the pixel from where the iterations started) are set to the final value (average color at the last iteration): \f[I(X,Y) <- (R*,G*,B*)\f] When maxLevel > 0, the gaussian pyramid of maxLevel+1 levels is built, and the above procedure is run on the smallest layer first. After that, the results are propagated to the larger layer and the iterations are run again only on those pixels where the layer colors differ by more than sr from the lower-resolution layer of the pyramid. That makes boundaries of color regions sharper. Note that the results will be actually different from the ones obtained by running the meanshift procedure on the whole original image (i.e. when maxLevel==0).

Python prototype (for reference only):

pyrMeanShiftFiltering(src, sp, sr[, dst[, maxLevel[, termcrit]]]) -> dst
@spec pyrUp(Evision.Mat.maybe_mat_in()) :: Evision.Mat.t() | {:error, String.t()}

Upsamples an image and then blurs it.

Positional Arguments
  • src: Evision.Mat.t().

    input image.

Keyword Arguments
  • dstsize: Size.

    size of the output image.

  • borderType: int.

    Pixel extrapolation method, see #BorderTypes (only #BORDER_DEFAULT is supported)

Return
  • dst: Evision.Mat.t().

    output image. It has the specified size and the same type as src .

By default, size of the output image is computed as Size(src.cols\*2, (src.rows\*2), but in any case, the following conditions should be satisfied: \f[\begin{array}{l} | \texttt{dstsize.width} -src.cols*2| \leq ( \texttt{dstsize.width} \mod 2) \\ | \texttt{dstsize.height} -src.rows*2| \leq ( \texttt{dstsize.height} \mod 2) \end{array}\f] The function performs the upsampling step of the Gaussian pyramid construction, though it can actually be used to construct the Laplacian pyramid. First, it upsamples the source image by injecting even zero rows and columns and then convolves the result with the same kernel as in pyrDown multiplied by 4.

Python prototype (for reference only):

pyrUp(src[, dst[, dstsize[, borderType]]]) -> dst
@spec pyrUp(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Upsamples an image and then blurs it.

Positional Arguments
  • src: Evision.Mat.t().

    input image.

Keyword Arguments
  • dstsize: Size.

    size of the output image.

  • borderType: int.

    Pixel extrapolation method, see #BorderTypes (only #BORDER_DEFAULT is supported)

Return
  • dst: Evision.Mat.t().

    output image. It has the specified size and the same type as src .

By default, size of the output image is computed as Size(src.cols\*2, (src.rows\*2), but in any case, the following conditions should be satisfied: \f[\begin{array}{l} | \texttt{dstsize.width} -src.cols*2| \leq ( \texttt{dstsize.width} \mod 2) \\ | \texttt{dstsize.height} -src.rows*2| \leq ( \texttt{dstsize.height} \mod 2) \end{array}\f] The function performs the upsampling step of the Gaussian pyramid construction, though it can actually be used to construct the Laplacian pyramid. First, it upsamples the source image by injecting even zero rows and columns and then convolves the result with the same kernel as in pyrDown multiplied by 4.

Python prototype (for reference only):

pyrUp(src[, dst[, dstsize[, borderType]]]) -> dst
Link to this function

randn(dst, mean, stddev)

View Source

Fills the array with normally distributed random numbers.

Positional Arguments
  • mean: Evision.Mat.t().

    mean value (expectation) of the generated random numbers.

  • stddev: Evision.Mat.t().

    standard deviation of the generated random numbers; it can be either a vector (in which case a diagonal standard deviation matrix is assumed) or a square matrix.

Return
  • dst: Evision.Mat.t().

    output array of random numbers; the array must be pre-allocated and have 1 to 4 channels.

The function cv::randn fills the matrix dst with normally distributed random numbers with the specified mean vector and the standard deviation matrix. The generated random numbers are clipped to fit the value range of the output array data type. @sa RNG, randu

Python prototype (for reference only):

randn(dst, mean, stddev) -> dst
@spec randShuffle(Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

Shuffles the array elements randomly.

Keyword Arguments
  • iterFactor: double.

    scale factor that determines the number of random swap operations (see the details below).

Return
  • dst: Evision.Mat.t().

    input/output numerical 1D array.

The function cv::randShuffle shuffles the specified 1D array by randomly choosing pairs of elements and swapping them. The number of such swap operations will be dst.rows*dst.cols*iterFactor . @sa RNG, sort

Python prototype (for reference only):

randShuffle(dst[, iterFactor]) -> dst
@spec randShuffle(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Shuffles the array elements randomly.

Keyword Arguments
  • iterFactor: double.

    scale factor that determines the number of random swap operations (see the details below).

Return
  • dst: Evision.Mat.t().

    input/output numerical 1D array.

The function cv::randShuffle shuffles the specified 1D array by randomly choosing pairs of elements and swapping them. The number of such swap operations will be dst.rows*dst.cols*iterFactor . @sa RNG, sort

Python prototype (for reference only):

randShuffle(dst[, iterFactor]) -> dst

Generates a single uniformly-distributed random number or an array of random numbers.

Positional Arguments
  • low: Evision.Mat.t().

    inclusive lower boundary of the generated random numbers.

  • high: Evision.Mat.t().

    exclusive upper boundary of the generated random numbers.

Return
  • dst: Evision.Mat.t().

    output array of random numbers; the array must be pre-allocated.

Non-template variant of the function fills the matrix dst with uniformly-distributed random numbers from the specified range: \f[\texttt{low} _c \leq \texttt{dst} (I)_c < \texttt{high} _c\f] @sa RNG, randn, theRNG

Python prototype (for reference only):

randu(dst, low, high) -> dst
@spec readOpticalFlow(binary()) :: Evision.Mat.t() | {:error, String.t()}

Read a .flo file

Positional Arguments
  • path: String.

    Path to the file to be loaded

Return
  • retval: Evision.Mat.t()

The function readOpticalFlow loads a flow field from a file and returns it as a single matrix. Resulting Mat has a type CV_32FC2 - floating-point, 2-channel. First channel corresponds to the flow in the horizontal direction (u), second - vertical (v).

Python prototype (for reference only):

readOpticalFlow(path) -> retval
Link to this function

recoverPose(e, points1, points2)

View Source

recoverPose

Positional Arguments
  • e: Evision.Mat.t().

    The input essential matrix.

  • points1: Evision.Mat.t().

    Array of N 2D points from the first image. The point coordinates should be floating-point (single or double precision).

  • points2: Evision.Mat.t().

    Array of the second image points of the same size and format as points1 .

Keyword Arguments
  • focal: double.

    Focal length of the camera. Note that this function assumes that points1 and points2 are feature points from cameras with same focal length and principal point.

  • pp: Point2d.

    principal point of the camera.

Return
  • retval: int

  • r: Evision.Mat.t().

    Output rotation matrix. Together with the translation vector, this matrix makes up a tuple that performs a change of basis from the first camera's coordinate system to the second camera's coordinate system. Note that, in general, t can not be used for this tuple, see the parameter description below.

  • t: Evision.Mat.t().

    Output translation vector. This vector is obtained by @ref decomposeEssentialMat and therefore is only known up to scale, i.e. t is the direction of the translation vector and has unit length.

  • mask: Evision.Mat.t().

    Input/output mask for inliers in points1 and points2. If it is not empty, then it marks inliers in points1 and points2 for the given essential matrix E. Only these inliers will be used to recover pose. In the output mask only inliers which pass the chirality check.

Has overloading in C++

This function differs from the one above that it computes camera intrinsic matrix from focal length and principal point: \f[A = \begin{bmatrix} f & 0 & x_{pp} \\ 0 & f & y_{pp} \\ 0 & 0 & 1 \end{bmatrix}\f]

Python prototype (for reference only):

recoverPose(E, points1, points2[, R[, t[, focal[, pp[, mask]]]]]) -> retval, R, t, mask
Link to this function

recoverPose(e, points1, points2, opts)

View Source

Variant 1:

Recovers the relative camera rotation and the translation from an estimated essential matrix and the corresponding points in two images, using chirality check. Returns the number of inliers that pass the check.

Positional Arguments
  • e: Evision.Mat.t().

    The input essential matrix.

  • points1: Evision.Mat.t().

    Array of N 2D points from the first image. The point coordinates should be floating-point (single or double precision).

  • points2: Evision.Mat.t().

    Array of the second image points of the same size and format as points1 .

  • cameraMatrix: Evision.Mat.t().

    Camera intrinsic matrix \f$\cameramatrix{A}\f$ . Note that this function assumes that points1 and points2 are feature points from cameras with the same camera intrinsic matrix.

Return
  • retval: int

  • r: Evision.Mat.t().

    Output rotation matrix. Together with the translation vector, this matrix makes up a tuple that performs a change of basis from the first camera's coordinate system to the second camera's coordinate system. Note that, in general, t can not be used for this tuple, see the parameter described below.

  • t: Evision.Mat.t().

    Output translation vector. This vector is obtained by @ref decomposeEssentialMat and therefore is only known up to scale, i.e. t is the direction of the translation vector and has unit length.

  • mask: Evision.Mat.t().

    Input/output mask for inliers in points1 and points2. If it is not empty, then it marks inliers in points1 and points2 for the given essential matrix E. Only these inliers will be used to recover pose. In the output mask only inliers which pass the chirality check.

This function decomposes an essential matrix using @ref decomposeEssentialMat and then verifies possible pose hypotheses by doing chirality check. The chirality check means that the triangulated 3D points should have positive depth. Some details can be found in @cite Nister03. This function can be used to process the output E and mask from @ref findEssentialMat. In this scenario, points1 and points2 are the same input for #findEssentialMat :

// Example. Estimation of fundamental matrix using the RANSAC algorithm
int point_count = 100;
vector<Point2f> points1(point_count);
vector<Point2f> points2(point_count);
// initialize the points here ...
for( int i = 0; i < point_count; i++ )
{
points1[i] = ...;
points2[i] = ...;
}
// cametra matrix with both focal lengths = 1, and principal point = (0, 0)
Mat cameraMatrix = Mat::eye(3, 3, CV_64F);
Mat E, R, t, mask;
E = findEssentialMat(points1, points2, cameraMatrix, RANSAC, 0.999, 1.0, mask);
recoverPose(E, points1, points2, cameraMatrix, R, t, mask);

Python prototype (for reference only):

recoverPose(E, points1, points2, cameraMatrix[, R[, t[, mask]]]) -> retval, R, t, mask

Variant 2:

recoverPose

Positional Arguments
  • e: Evision.Mat.t().

    The input essential matrix.

  • points1: Evision.Mat.t().

    Array of N 2D points from the first image. The point coordinates should be floating-point (single or double precision).

  • points2: Evision.Mat.t().

    Array of the second image points of the same size and format as points1 .

Keyword Arguments
  • focal: double.

    Focal length of the camera. Note that this function assumes that points1 and points2 are feature points from cameras with same focal length and principal point.

  • pp: Point2d.

    principal point of the camera.

Return
  • retval: int

  • r: Evision.Mat.t().

    Output rotation matrix. Together with the translation vector, this matrix makes up a tuple that performs a change of basis from the first camera's coordinate system to the second camera's coordinate system. Note that, in general, t can not be used for this tuple, see the parameter description below.

  • t: Evision.Mat.t().

    Output translation vector. This vector is obtained by @ref decomposeEssentialMat and therefore is only known up to scale, i.e. t is the direction of the translation vector and has unit length.

  • mask: Evision.Mat.t().

    Input/output mask for inliers in points1 and points2. If it is not empty, then it marks inliers in points1 and points2 for the given essential matrix E. Only these inliers will be used to recover pose. In the output mask only inliers which pass the chirality check.

Has overloading in C++

This function differs from the one above that it computes camera intrinsic matrix from focal length and principal point: \f[A = \begin{bmatrix} f & 0 & x_{pp} \\ 0 & f & y_{pp} \\ 0 & 0 & 1 \end{bmatrix}\f]

Python prototype (for reference only):

recoverPose(E, points1, points2[, R[, t[, focal[, pp[, mask]]]]]) -> retval, R, t, mask
Link to this function

recoverPose(e, points1, points2, cameraMatrix, opts)

View Source

Variant 1:

recoverPose

Positional Arguments
  • e: Evision.Mat.t().

    The input essential matrix.

  • points1: Evision.Mat.t().

    Array of N 2D points from the first image. The point coordinates should be floating-point (single or double precision).

  • points2: Evision.Mat.t().

    Array of the second image points of the same size and format as points1.

  • cameraMatrix: Evision.Mat.t().

    Camera intrinsic matrix \f$\cameramatrix{A}\f$ . Note that this function assumes that points1 and points2 are feature points from cameras with the same camera intrinsic matrix.

  • distanceThresh: double.

    threshold distance which is used to filter out far away points (i.e. infinite points).

Return
  • retval: int

  • r: Evision.Mat.t().

    Output rotation matrix. Together with the translation vector, this matrix makes up a tuple that performs a change of basis from the first camera's coordinate system to the second camera's coordinate system. Note that, in general, t can not be used for this tuple, see the parameter description below.

  • t: Evision.Mat.t().

    Output translation vector. This vector is obtained by @ref decomposeEssentialMat and therefore is only known up to scale, i.e. t is the direction of the translation vector and has unit length.

  • mask: Evision.Mat.t().

    Input/output mask for inliers in points1 and points2. If it is not empty, then it marks inliers in points1 and points2 for the given essential matrix E. Only these inliers will be used to recover pose. In the output mask only inliers which pass the chirality check.

  • triangulatedPoints: Evision.Mat.t().

    3D points which were reconstructed by triangulation.

Has overloading in C++

This function differs from the one above that it outputs the triangulated 3D point that are used for the chirality check.

Python prototype (for reference only):

recoverPose(E, points1, points2, cameraMatrix, distanceThresh[, R[, t[, mask[, triangulatedPoints]]]]) -> retval, R, t, mask, triangulatedPoints

Variant 2:

Recovers the relative camera rotation and the translation from an estimated essential matrix and the corresponding points in two images, using chirality check. Returns the number of inliers that pass the check.

Positional Arguments
  • e: Evision.Mat.t().

    The input essential matrix.

  • points1: Evision.Mat.t().

    Array of N 2D points from the first image. The point coordinates should be floating-point (single or double precision).

  • points2: Evision.Mat.t().

    Array of the second image points of the same size and format as points1 .

  • cameraMatrix: Evision.Mat.t().

    Camera intrinsic matrix \f$\cameramatrix{A}\f$ . Note that this function assumes that points1 and points2 are feature points from cameras with the same camera intrinsic matrix.

Return
  • retval: int

  • r: Evision.Mat.t().

    Output rotation matrix. Together with the translation vector, this matrix makes up a tuple that performs a change of basis from the first camera's coordinate system to the second camera's coordinate system. Note that, in general, t can not be used for this tuple, see the parameter described below.

  • t: Evision.Mat.t().

    Output translation vector. This vector is obtained by @ref decomposeEssentialMat and therefore is only known up to scale, i.e. t is the direction of the translation vector and has unit length.

  • mask: Evision.Mat.t().

    Input/output mask for inliers in points1 and points2. If it is not empty, then it marks inliers in points1 and points2 for the given essential matrix E. Only these inliers will be used to recover pose. In the output mask only inliers which pass the chirality check.

This function decomposes an essential matrix using @ref decomposeEssentialMat and then verifies possible pose hypotheses by doing chirality check. The chirality check means that the triangulated 3D points should have positive depth. Some details can be found in @cite Nister03. This function can be used to process the output E and mask from @ref findEssentialMat. In this scenario, points1 and points2 are the same input for #findEssentialMat :

// Example. Estimation of fundamental matrix using the RANSAC algorithm
int point_count = 100;
vector<Point2f> points1(point_count);
vector<Point2f> points2(point_count);
// initialize the points here ...
for( int i = 0; i < point_count; i++ )
{
points1[i] = ...;
points2[i] = ...;
}
// cametra matrix with both focal lengths = 1, and principal point = (0, 0)
Mat cameraMatrix = Mat::eye(3, 3, CV_64F);
Mat E, R, t, mask;
E = findEssentialMat(points1, points2, cameraMatrix, RANSAC, 0.999, 1.0, mask);
recoverPose(E, points1, points2, cameraMatrix, R, t, mask);

Python prototype (for reference only):

recoverPose(E, points1, points2, cameraMatrix[, R[, t[, mask]]]) -> retval, R, t, mask
Link to this function

recoverPose(e, points1, points2, cameraMatrix, distanceThresh, opts)

View Source

Variant 1:

Recovers the relative camera rotation and the translation from corresponding points in two images from two different cameras, using cheirality check. Returns the number of inliers that pass the check.

Positional Arguments
  • points1: Evision.Mat.t().

    Array of N 2D points from the first image. The point coordinates should be floating-point (single or double precision).

  • points2: Evision.Mat.t().

    Array of the second image points of the same size and format as points1 .

  • cameraMatrix1: Evision.Mat.t().

    Input/output camera matrix for the first camera, the same as in

  • distCoeffs1: Evision.Mat.t().

    Input/output vector of distortion coefficients, the same as in

  • cameraMatrix2: Evision.Mat.t().

    Input/output camera matrix for the first camera, the same as in

  • distCoeffs2: Evision.Mat.t().

    Input/output vector of distortion coefficients, the same as in

Keyword Arguments
  • method: int.

    Method for computing an essential matrix.

    • @ref RANSAC for the RANSAC algorithm.
    • @ref LMEDS for the LMedS algorithm.
  • prob: double.

    Parameter used for the RANSAC or LMedS methods only. It specifies a desirable level of confidence (probability) that the estimated matrix is correct.

  • threshold: double.

    Parameter used for RANSAC. It is the maximum distance from a point to an epipolar line in pixels, beyond which the point is considered an outlier and is not used for computing the final fundamental matrix. It can be set to something like 1-3, depending on the accuracy of the point localization, image resolution, and the image noise.

Return
  • retval: int

  • e: Evision.Mat.t().

    The output essential matrix.

  • r: Evision.Mat.t().

    Output rotation matrix. Together with the translation vector, this matrix makes up a tuple that performs a change of basis from the first camera's coordinate system to the second camera's coordinate system. Note that, in general, t can not be used for this tuple, see the parameter described below.

  • t: Evision.Mat.t().

    Output translation vector. This vector is obtained by @ref decomposeEssentialMat and therefore is only known up to scale, i.e. t is the direction of the translation vector and has unit length.

  • mask: Evision.Mat.t().

    Input/output mask for inliers in points1 and points2. If it is not empty, then it marks inliers in points1 and points2 for then given essential matrix E. Only these inliers will be used to recover pose. In the output mask only inliers which pass the cheirality check.

@ref calibrateCamera. Furthermore, for the stereo case, additional flags may be used, see below. @ref calibrateCamera. @ref calibrateCamera. Furthermore, for the stereo case, additional flags may be used, see below. @ref calibrateCamera.

This function decomposes an essential matrix using @ref decomposeEssentialMat and then verifies possible pose hypotheses by doing cheirality check. The cheirality check means that the triangulated 3D points should have positive depth. Some details can be found in @cite Nister03. This function can be used to process the output E and mask from @ref findEssentialMat. In this scenario, points1 and points2 are the same input for findEssentialMat.:

// Example. Estimation of fundamental matrix using the RANSAC algorithm
int point_count = 100;
vector<Point2f> points1(point_count);
vector<Point2f> points2(point_count);
// initialize the points here ...
for( int i = 0; i < point_count; i++ )
{
points1[i] = ...;
points2[i] = ...;
}
// Input: camera calibration of both cameras, for example using intrinsic chessboard calibration.
Mat cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2;
// Output: Essential matrix, relative rotation and relative translation.
Mat E, R, t, mask;
recoverPose(points1, points2, cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2, E, R, t, mask);

Python prototype (for reference only):

recoverPose(points1, points2, cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2[, E[, R[, t[, method[, prob[, threshold[, mask]]]]]]]) -> retval, E, R, t, mask

Variant 2:

recoverPose

Positional Arguments
  • e: Evision.Mat.t().

    The input essential matrix.

  • points1: Evision.Mat.t().

    Array of N 2D points from the first image. The point coordinates should be floating-point (single or double precision).

  • points2: Evision.Mat.t().

    Array of the second image points of the same size and format as points1.

  • cameraMatrix: Evision.Mat.t().

    Camera intrinsic matrix \f$\cameramatrix{A}\f$ . Note that this function assumes that points1 and points2 are feature points from cameras with the same camera intrinsic matrix.

  • distanceThresh: double.

    threshold distance which is used to filter out far away points (i.e. infinite points).

Return
  • retval: int

  • r: Evision.Mat.t().

    Output rotation matrix. Together with the translation vector, this matrix makes up a tuple that performs a change of basis from the first camera's coordinate system to the second camera's coordinate system. Note that, in general, t can not be used for this tuple, see the parameter description below.

  • t: Evision.Mat.t().

    Output translation vector. This vector is obtained by @ref decomposeEssentialMat and therefore is only known up to scale, i.e. t is the direction of the translation vector and has unit length.

  • mask: Evision.Mat.t().

    Input/output mask for inliers in points1 and points2. If it is not empty, then it marks inliers in points1 and points2 for the given essential matrix E. Only these inliers will be used to recover pose. In the output mask only inliers which pass the chirality check.

  • triangulatedPoints: Evision.Mat.t().

    3D points which were reconstructed by triangulation.

Has overloading in C++

This function differs from the one above that it outputs the triangulated 3D point that are used for the chirality check.

Python prototype (for reference only):

recoverPose(E, points1, points2, cameraMatrix, distanceThresh[, R[, t[, mask[, triangulatedPoints]]]]) -> retval, R, t, mask, triangulatedPoints
Link to this function

recoverPose(points1, points2, cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2, opts)

View Source

Recovers the relative camera rotation and the translation from corresponding points in two images from two different cameras, using cheirality check. Returns the number of inliers that pass the check.

Positional Arguments
  • points1: Evision.Mat.t().

    Array of N 2D points from the first image. The point coordinates should be floating-point (single or double precision).

  • points2: Evision.Mat.t().

    Array of the second image points of the same size and format as points1 .

  • cameraMatrix1: Evision.Mat.t().

    Input/output camera matrix for the first camera, the same as in

  • distCoeffs1: Evision.Mat.t().

    Input/output vector of distortion coefficients, the same as in

  • cameraMatrix2: Evision.Mat.t().

    Input/output camera matrix for the first camera, the same as in

  • distCoeffs2: Evision.Mat.t().

    Input/output vector of distortion coefficients, the same as in

Keyword Arguments
  • method: int.

    Method for computing an essential matrix.

    • @ref RANSAC for the RANSAC algorithm.
    • @ref LMEDS for the LMedS algorithm.
  • prob: double.

    Parameter used for the RANSAC or LMedS methods only. It specifies a desirable level of confidence (probability) that the estimated matrix is correct.

  • threshold: double.

    Parameter used for RANSAC. It is the maximum distance from a point to an epipolar line in pixels, beyond which the point is considered an outlier and is not used for computing the final fundamental matrix. It can be set to something like 1-3, depending on the accuracy of the point localization, image resolution, and the image noise.

Return
  • retval: int

  • e: Evision.Mat.t().

    The output essential matrix.

  • r: Evision.Mat.t().

    Output rotation matrix. Together with the translation vector, this matrix makes up a tuple that performs a change of basis from the first camera's coordinate system to the second camera's coordinate system. Note that, in general, t can not be used for this tuple, see the parameter described below.

  • t: Evision.Mat.t().

    Output translation vector. This vector is obtained by @ref decomposeEssentialMat and therefore is only known up to scale, i.e. t is the direction of the translation vector and has unit length.

  • mask: Evision.Mat.t().

    Input/output mask for inliers in points1 and points2. If it is not empty, then it marks inliers in points1 and points2 for then given essential matrix E. Only these inliers will be used to recover pose. In the output mask only inliers which pass the cheirality check.

@ref calibrateCamera. Furthermore, for the stereo case, additional flags may be used, see below. @ref calibrateCamera. @ref calibrateCamera. Furthermore, for the stereo case, additional flags may be used, see below. @ref calibrateCamera.

This function decomposes an essential matrix using @ref decomposeEssentialMat and then verifies possible pose hypotheses by doing cheirality check. The cheirality check means that the triangulated 3D points should have positive depth. Some details can be found in @cite Nister03. This function can be used to process the output E and mask from @ref findEssentialMat. In this scenario, points1 and points2 are the same input for findEssentialMat.:

// Example. Estimation of fundamental matrix using the RANSAC algorithm
int point_count = 100;
vector<Point2f> points1(point_count);
vector<Point2f> points2(point_count);
// initialize the points here ...
for( int i = 0; i < point_count; i++ )
{
points1[i] = ...;
points2[i] = ...;
}
// Input: camera calibration of both cameras, for example using intrinsic chessboard calibration.
Mat cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2;
// Output: Essential matrix, relative rotation and relative translation.
Mat E, R, t, mask;
recoverPose(points1, points2, cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2, E, R, t, mask);

Python prototype (for reference only):

recoverPose(points1, points2, cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2[, E[, R[, t[, method[, prob[, threshold[, mask]]]]]]]) -> retval, E, R, t, mask
Link to this function

rectangle(img, rec, color)

View Source
@spec rectangle(
  Evision.Mat.maybe_mat_in(),
  {number(), number(), number(), number()},
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()}
) :: Evision.Mat.t() | {:error, String.t()}

rectangle

Positional Arguments
  • rec: Rect
  • color: Scalar
Keyword Arguments
  • thickness: int.
  • lineType: int.
  • shift: int.
Return
  • img: Evision.Mat.t()

Has overloading in C++

use rec parameter as alternative specification of the drawn rectangle: r.tl() and r.br()-Point(1,1) are opposite corners

Python prototype (for reference only):

rectangle(img, rec, color[, thickness[, lineType[, shift]]]) -> img
Link to this function

rectangle(img, rec, color, opts)

View Source
@spec rectangle(
  Evision.Mat.maybe_mat_in(),
  {number(), number(), number(), number()},
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()},
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}
@spec rectangle(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  {number(), number()},
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()}
) :: Evision.Mat.t() | {:error, String.t()}

Variant 1:

Draws a simple, thick, or filled up-right rectangle.

Positional Arguments
  • pt1: Point.

    Vertex of the rectangle.

  • pt2: Point.

    Vertex of the rectangle opposite to pt1 .

  • color: Scalar.

    Rectangle color or brightness (grayscale image).

Keyword Arguments
  • thickness: int.

    Thickness of lines that make up the rectangle. Negative values, like #FILLED, mean that the function has to draw a filled rectangle.

  • lineType: int.

    Type of the line. See #LineTypes

  • shift: int.

    Number of fractional bits in the point coordinates.

Return
  • img: Evision.Mat.t().

    Image.

The function cv::rectangle draws a rectangle outline or a filled rectangle whose two opposite corners are pt1 and pt2.

Python prototype (for reference only):

rectangle(img, pt1, pt2, color[, thickness[, lineType[, shift]]]) -> img

Variant 2:

rectangle

Positional Arguments
  • rec: Rect
  • color: Scalar
Keyword Arguments
  • thickness: int.
  • lineType: int.
  • shift: int.
Return
  • img: Evision.Mat.t()

Has overloading in C++

use rec parameter as alternative specification of the drawn rectangle: r.tl() and r.br()-Point(1,1) are opposite corners

Python prototype (for reference only):

rectangle(img, rec, color[, thickness[, lineType[, shift]]]) -> img
Link to this function

rectangle(img, pt1, pt2, color, opts)

View Source
@spec rectangle(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  {number(), number()},
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()},
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Draws a simple, thick, or filled up-right rectangle.

Positional Arguments
  • pt1: Point.

    Vertex of the rectangle.

  • pt2: Point.

    Vertex of the rectangle opposite to pt1 .

  • color: Scalar.

    Rectangle color or brightness (grayscale image).

Keyword Arguments
  • thickness: int.

    Thickness of lines that make up the rectangle. Negative values, like #FILLED, mean that the function has to draw a filled rectangle.

  • lineType: int.

    Type of the line. See #LineTypes

  • shift: int.

    Number of fractional bits in the point coordinates.

Return
  • img: Evision.Mat.t().

    Image.

The function cv::rectangle draws a rectangle outline or a filled rectangle whose two opposite corners are pt1 and pt2.

Python prototype (for reference only):

rectangle(img, pt1, pt2, color[, thickness[, lineType[, shift]]]) -> img
Link to this function

rectangleIntersectionArea(a, b)

View Source
@spec rectangleIntersectionArea(
  {number(), number(), number(), number()},
  {number(), number(), number(), number()}
) :: number() | {:error, String.t()}

Finds out if there is any intersection between two rectangles

Positional Arguments
  • a: Rect2d.

    First rectangle

  • b: Rect2d.

    Second rectangle

Return
  • retval: double

mainly useful for language bindings @return the area of the intersection

Python prototype (for reference only):

rectangleIntersectionArea(a, b) -> retval
Link to this function

rectify3Collinear(cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2, cameraMatrix3, distCoeffs3, imgpt1, imgpt3, imageSize, r12, t12, r13, t13, alpha, newImgSize, flags)

View Source

rectify3Collinear

Positional Arguments
  • cameraMatrix1: Evision.Mat.t()
  • distCoeffs1: Evision.Mat.t()
  • cameraMatrix2: Evision.Mat.t()
  • distCoeffs2: Evision.Mat.t()
  • cameraMatrix3: Evision.Mat.t()
  • distCoeffs3: Evision.Mat.t()
  • imgpt1: [Evision.Mat]
  • imgpt3: [Evision.Mat]
  • imageSize: Size
  • r12: Evision.Mat.t()
  • t12: Evision.Mat.t()
  • r13: Evision.Mat.t()
  • t13: Evision.Mat.t()
  • alpha: double
  • newImgSize: Size
  • flags: int
Return
  • retval: float
  • r1: Evision.Mat.t().
  • r2: Evision.Mat.t().
  • r3: Evision.Mat.t().
  • p1: Evision.Mat.t().
  • p2: Evision.Mat.t().
  • p3: Evision.Mat.t().
  • q: Evision.Mat.t().
  • roi1: Rect*
  • roi2: Rect*

Python prototype (for reference only):

rectify3Collinear(cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2, cameraMatrix3, distCoeffs3, imgpt1, imgpt3, imageSize, R12, T12, R13, T13, alpha, newImgSize, flags[, R1[, R2[, R3[, P1[, P2[, P3[, Q]]]]]]]) -> retval, R1, R2, R3, P1, P2, P3, Q, roi1, roi2
Link to this function

rectify3Collinear(cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2, cameraMatrix3, distCoeffs3, imgpt1, imgpt3, imageSize, r12, t12, r13, t13, alpha, newImgSize, flags, opts)

View Source

rectify3Collinear

Positional Arguments
  • cameraMatrix1: Evision.Mat.t()
  • distCoeffs1: Evision.Mat.t()
  • cameraMatrix2: Evision.Mat.t()
  • distCoeffs2: Evision.Mat.t()
  • cameraMatrix3: Evision.Mat.t()
  • distCoeffs3: Evision.Mat.t()
  • imgpt1: [Evision.Mat]
  • imgpt3: [Evision.Mat]
  • imageSize: Size
  • r12: Evision.Mat.t()
  • t12: Evision.Mat.t()
  • r13: Evision.Mat.t()
  • t13: Evision.Mat.t()
  • alpha: double
  • newImgSize: Size
  • flags: int
Return
  • retval: float
  • r1: Evision.Mat.t().
  • r2: Evision.Mat.t().
  • r3: Evision.Mat.t().
  • p1: Evision.Mat.t().
  • p2: Evision.Mat.t().
  • p3: Evision.Mat.t().
  • q: Evision.Mat.t().
  • roi1: Rect*
  • roi2: Rect*

Python prototype (for reference only):

rectify3Collinear(cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2, cameraMatrix3, distCoeffs3, imgpt1, imgpt3, imageSize, R12, T12, R13, T13, alpha, newImgSize, flags[, R1[, R2[, R3[, P1[, P2[, P3[, Q]]]]]]]) -> retval, R1, R2, R3, P1, P2, P3, Q, roi1, roi2
@spec reduce(Evision.Mat.maybe_mat_in(), integer(), integer()) ::
  Evision.Mat.t() | {:error, String.t()}

Reduces a matrix to a vector.

Positional Arguments
  • src: Evision.Mat.t().

    input 2D matrix.

  • dim: int.

    dimension index along which the matrix is reduced. 0 means that the matrix is reduced to a single row. 1 means that the matrix is reduced to a single column.

  • rtype: int.

    reduction operation that could be one of #ReduceTypes

Keyword Arguments
  • dtype: int.

    when negative, the output vector will have the same type as the input matrix, otherwise, its type will be CV_MAKE_TYPE(CV_MAT_DEPTH(dtype), src.channels()).

Return
  • dst: Evision.Mat.t().

    output vector. Its size and type is defined by dim and dtype parameters.

The function #reduce reduces the matrix to a vector by treating the matrix rows/columns as a set of 1D vectors and performing the specified operation on the vectors until a single row/column is obtained. For example, the function can be used to compute horizontal and vertical projections of a raster image. In case of #REDUCE_MAX and #REDUCE_MIN , the output image should have the same type as the source one. In case of #REDUCE_SUM, #REDUCE_SUM2 and #REDUCE_AVG , the output may have a larger element bit-depth to preserve accuracy. And multi-channel arrays are also supported in these two reduction modes. The following code demonstrates its usage for a single channel matrix. @snippet snippets/core_reduce.cpp example And the following code demonstrates its usage for a two-channel matrix. @snippet snippets/core_reduce.cpp example2 @sa repeat, reduceArgMin, reduceArgMax

Python prototype (for reference only):

reduce(src, dim, rtype[, dst[, dtype]]) -> dst
Link to this function

reduce(src, dim, rtype, opts)

View Source
@spec reduce(
  Evision.Mat.maybe_mat_in(),
  integer(),
  integer(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Reduces a matrix to a vector.

Positional Arguments
  • src: Evision.Mat.t().

    input 2D matrix.

  • dim: int.

    dimension index along which the matrix is reduced. 0 means that the matrix is reduced to a single row. 1 means that the matrix is reduced to a single column.

  • rtype: int.

    reduction operation that could be one of #ReduceTypes

Keyword Arguments
  • dtype: int.

    when negative, the output vector will have the same type as the input matrix, otherwise, its type will be CV_MAKE_TYPE(CV_MAT_DEPTH(dtype), src.channels()).

Return
  • dst: Evision.Mat.t().

    output vector. Its size and type is defined by dim and dtype parameters.

The function #reduce reduces the matrix to a vector by treating the matrix rows/columns as a set of 1D vectors and performing the specified operation on the vectors until a single row/column is obtained. For example, the function can be used to compute horizontal and vertical projections of a raster image. In case of #REDUCE_MAX and #REDUCE_MIN , the output image should have the same type as the source one. In case of #REDUCE_SUM, #REDUCE_SUM2 and #REDUCE_AVG , the output may have a larger element bit-depth to preserve accuracy. And multi-channel arrays are also supported in these two reduction modes. The following code demonstrates its usage for a single channel matrix. @snippet snippets/core_reduce.cpp example And the following code demonstrates its usage for a two-channel matrix. @snippet snippets/core_reduce.cpp example2 @sa repeat, reduceArgMin, reduceArgMax

Python prototype (for reference only):

reduce(src, dim, rtype[, dst[, dtype]]) -> dst
@spec reduceArgMax(Evision.Mat.maybe_mat_in(), integer()) ::
  Evision.Mat.t() | {:error, String.t()}

Finds indices of max elements along provided axis

Positional Arguments
  • src: Evision.Mat.t().

    input single-channel array.

  • axis: int.

    axis to reduce along.

Keyword Arguments
  • lastIndex: bool.

    whether to get the index of first or last occurrence of max.

Return
  • dst: Evision.Mat.t().

    output array of type CV_32SC1 with the same dimensionality as src, except for axis being reduced - it should be set to 1.

Note:

  • If input or output array is not continuous, this function will create an internal copy.
  • NaN handling is left unspecified, see patchNaNs().
  • The returned index is always in bounds of input matrix.

@sa reduceArgMin, minMaxLoc, min, max, compare, reduce

Python prototype (for reference only):

reduceArgMax(src, axis[, dst[, lastIndex]]) -> dst
Link to this function

reduceArgMax(src, axis, opts)

View Source
@spec reduceArgMax(
  Evision.Mat.maybe_mat_in(),
  integer(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Finds indices of max elements along provided axis

Positional Arguments
  • src: Evision.Mat.t().

    input single-channel array.

  • axis: int.

    axis to reduce along.

Keyword Arguments
  • lastIndex: bool.

    whether to get the index of first or last occurrence of max.

Return
  • dst: Evision.Mat.t().

    output array of type CV_32SC1 with the same dimensionality as src, except for axis being reduced - it should be set to 1.

Note:

  • If input or output array is not continuous, this function will create an internal copy.
  • NaN handling is left unspecified, see patchNaNs().
  • The returned index is always in bounds of input matrix.

@sa reduceArgMin, minMaxLoc, min, max, compare, reduce

Python prototype (for reference only):

reduceArgMax(src, axis[, dst[, lastIndex]]) -> dst
@spec reduceArgMin(Evision.Mat.maybe_mat_in(), integer()) ::
  Evision.Mat.t() | {:error, String.t()}

Finds indices of min elements along provided axis

Positional Arguments
  • src: Evision.Mat.t().

    input single-channel array.

  • axis: int.

    axis to reduce along.

Keyword Arguments
  • lastIndex: bool.

    whether to get the index of first or last occurrence of min.

Return
  • dst: Evision.Mat.t().

    output array of type CV_32SC1 with the same dimensionality as src, except for axis being reduced - it should be set to 1.

Note:

  • If input or output array is not continuous, this function will create an internal copy.
  • NaN handling is left unspecified, see patchNaNs().
  • The returned index is always in bounds of input matrix.

@sa reduceArgMax, minMaxLoc, min, max, compare, reduce

Python prototype (for reference only):

reduceArgMin(src, axis[, dst[, lastIndex]]) -> dst
Link to this function

reduceArgMin(src, axis, opts)

View Source
@spec reduceArgMin(
  Evision.Mat.maybe_mat_in(),
  integer(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Finds indices of min elements along provided axis

Positional Arguments
  • src: Evision.Mat.t().

    input single-channel array.

  • axis: int.

    axis to reduce along.

Keyword Arguments
  • lastIndex: bool.

    whether to get the index of first or last occurrence of min.

Return
  • dst: Evision.Mat.t().

    output array of type CV_32SC1 with the same dimensionality as src, except for axis being reduced - it should be set to 1.

Note:

  • If input or output array is not continuous, this function will create an internal copy.
  • NaN handling is left unspecified, see patchNaNs().
  • The returned index is always in bounds of input matrix.

@sa reduceArgMax, minMaxLoc, min, max, compare, reduce

Python prototype (for reference only):

reduceArgMin(src, axis[, dst[, lastIndex]]) -> dst
Link to this function

remap(src, map1, map2, interpolation)

View Source

Applies a generic geometrical transformation to an image.

Positional Arguments
  • src: Evision.Mat.t().

    Source image.

  • map1: Evision.Mat.t().

    The first map of either (x,y) points or just x values having the type CV_16SC2 , CV_32FC1, or CV_32FC2. See #convertMaps for details on converting a floating point representation to fixed-point for speed.

  • map2: Evision.Mat.t().

    The second map of y values having the type CV_16UC1, CV_32FC1, or none (empty map if map1 is (x,y) points), respectively.

  • interpolation: int.

    Interpolation method (see #InterpolationFlags). The methods #INTER_AREA and #INTER_LINEAR_EXACT are not supported by this function.

Keyword Arguments
  • borderMode: int.

    Pixel extrapolation method (see #BorderTypes). When borderMode=#BORDER_TRANSPARENT, it means that the pixels in the destination image that corresponds to the "outliers" in the source image are not modified by the function.

  • borderValue: Scalar.

    Value used in case of a constant border. By default, it is 0.

Return
  • dst: Evision.Mat.t().

    Destination image. It has the same size as map1 and the same type as src .

The function remap transforms the source image using the specified map: \f[\texttt{dst} (x,y) = \texttt{src} (map_x(x,y),map_y(x,y))\f] where values of pixels with non-integer coordinates are computed using one of available interpolation methods. \f$map_x\f$ and \f$map_y\f$ can be encoded as separate floating-point maps in \f$map_1\f$ and \f$map_2\f$ respectively, or interleaved floating-point maps of \f$(x,y)\f$ in \f$map_1\f$, or fixed-point maps created by using #convertMaps. The reason you might want to convert from floating to fixed-point representations of a map is that they can yield much faster (\~2x) remapping operations. In the converted case, \f$map_1\f$ contains pairs (cvFloor(x), cvFloor(y)) and \f$map_2\f$ contains indices in a table of interpolation coefficients. This function cannot operate in-place. Note: Due to current implementation limitations the size of an input and output images should be less than 32767x32767.

Python prototype (for reference only):

remap(src, map1, map2, interpolation[, dst[, borderMode[, borderValue]]]) -> dst
Link to this function

remap(src, map1, map2, interpolation, opts)

View Source
@spec remap(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  integer(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Applies a generic geometrical transformation to an image.

Positional Arguments
  • src: Evision.Mat.t().

    Source image.

  • map1: Evision.Mat.t().

    The first map of either (x,y) points or just x values having the type CV_16SC2 , CV_32FC1, or CV_32FC2. See #convertMaps for details on converting a floating point representation to fixed-point for speed.

  • map2: Evision.Mat.t().

    The second map of y values having the type CV_16UC1, CV_32FC1, or none (empty map if map1 is (x,y) points), respectively.

  • interpolation: int.

    Interpolation method (see #InterpolationFlags). The methods #INTER_AREA and #INTER_LINEAR_EXACT are not supported by this function.

Keyword Arguments
  • borderMode: int.

    Pixel extrapolation method (see #BorderTypes). When borderMode=#BORDER_TRANSPARENT, it means that the pixels in the destination image that corresponds to the "outliers" in the source image are not modified by the function.

  • borderValue: Scalar.

    Value used in case of a constant border. By default, it is 0.

Return
  • dst: Evision.Mat.t().

    Destination image. It has the same size as map1 and the same type as src .

The function remap transforms the source image using the specified map: \f[\texttt{dst} (x,y) = \texttt{src} (map_x(x,y),map_y(x,y))\f] where values of pixels with non-integer coordinates are computed using one of available interpolation methods. \f$map_x\f$ and \f$map_y\f$ can be encoded as separate floating-point maps in \f$map_1\f$ and \f$map_2\f$ respectively, or interleaved floating-point maps of \f$(x,y)\f$ in \f$map_1\f$, or fixed-point maps created by using #convertMaps. The reason you might want to convert from floating to fixed-point representations of a map is that they can yield much faster (\~2x) remapping operations. In the converted case, \f$map_1\f$ contains pairs (cvFloor(x), cvFloor(y)) and \f$map_2\f$ contains indices in a table of interpolation coefficients. This function cannot operate in-place. Note: Due to current implementation limitations the size of an input and output images should be less than 32767x32767.

Python prototype (for reference only):

remap(src, map1, map2, interpolation[, dst[, borderMode[, borderValue]]]) -> dst
@spec repeat(Evision.Mat.maybe_mat_in(), integer(), integer()) ::
  Evision.Mat.t() | {:error, String.t()}

Fills the output array with repeated copies of the input array.

Positional Arguments
  • src: Evision.Mat.t().

    input array to replicate.

  • ny: int.

    Flag to specify how many times the src is repeated along the vertical axis.

  • nx: int.

    Flag to specify how many times the src is repeated along the horizontal axis.

Return
  • dst: Evision.Mat.t().

    output array of the same type as src.

The function cv::repeat duplicates the input array one or more times along each of the two axes: \f[\texttt{dst} _{ij}= \texttt{src} _{i\mod src.rows, \; j\mod src.cols }\f] The second variant of the function is more convenient to use with @ref MatrixExpressions. @sa cv::reduce

Python prototype (for reference only):

repeat(src, ny, nx[, dst]) -> dst
Link to this function

repeat(src, ny, nx, opts)

View Source
@spec repeat(
  Evision.Mat.maybe_mat_in(),
  integer(),
  integer(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Fills the output array with repeated copies of the input array.

Positional Arguments
  • src: Evision.Mat.t().

    input array to replicate.

  • ny: int.

    Flag to specify how many times the src is repeated along the vertical axis.

  • nx: int.

    Flag to specify how many times the src is repeated along the horizontal axis.

Return
  • dst: Evision.Mat.t().

    output array of the same type as src.

The function cv::repeat duplicates the input array one or more times along each of the two axes: \f[\texttt{dst} _{ij}= \texttt{src} _{i\mod src.rows, \; j\mod src.cols }\f] The second variant of the function is more convenient to use with @ref MatrixExpressions. @sa cv::reduce

Python prototype (for reference only):

repeat(src, ny, nx[, dst]) -> dst
Link to this function

reprojectImageTo3D(disparity, q)

View Source
@spec reprojectImageTo3D(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

Reprojects a disparity image to 3D space.

Positional Arguments
  • disparity: Evision.Mat.t().

    Input single-channel 8-bit unsigned, 16-bit signed, 32-bit signed or 32-bit floating-point disparity image. The values of 8-bit / 16-bit signed formats are assumed to have no fractional bits. If the disparity is 16-bit signed format, as computed by @ref StereoBM or

  • q: Evision.Mat.t().

    \f$4 \times 4\f$ perspective transformation matrix that can be obtained with

Keyword Arguments
  • handleMissingValues: bool.

    Indicates, whether the function should handle missing values (i.e. points where the disparity was not computed). If handleMissingValues=true, then pixels with the minimal disparity that corresponds to the outliers (see StereoMatcher::compute ) are transformed to 3D points with a very large Z value (currently set to 10000).

  • ddepth: int.

    The optional output array depth. If it is -1, the output image will have CV_32F depth. ddepth can also be set to CV_16S, CV_32S or CV_32F.

Return
  • 3dImage: Evision.Mat.t().

    Output 3-channel floating-point image of the same size as disparity. Each element of _3dImage(x,y) contains 3D coordinates of the point (x,y) computed from the disparity map. If one uses Q obtained by @ref stereoRectify, then the returned points are represented in the first camera's rectified coordinate system.

@ref StereoSGBM and maybe other algorithms, it should be divided by 16 (and scaled to float) before being used here. @ref stereoRectify.

The function transforms a single-channel disparity map to a 3-channel image representing a 3D surface. That is, for each pixel (x,y) and the corresponding disparity d=disparity(x,y) , it computes: \f[\begin{bmatrix} X \\ Y \\ Z \\ W \end{bmatrix} = Q \begin{bmatrix} x \\ y \\ \texttt{disparity} (x,y) \\ z \end{bmatrix}.\f] @sa To reproject a sparse set of points {(x,y,d),...} to 3D space, use perspectiveTransform.

Python prototype (for reference only):

reprojectImageTo3D(disparity, Q[, _3dImage[, handleMissingValues[, ddepth]]]) -> _3dImage
Link to this function

reprojectImageTo3D(disparity, q, opts)

View Source
@spec reprojectImageTo3D(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Reprojects a disparity image to 3D space.

Positional Arguments
  • disparity: Evision.Mat.t().

    Input single-channel 8-bit unsigned, 16-bit signed, 32-bit signed or 32-bit floating-point disparity image. The values of 8-bit / 16-bit signed formats are assumed to have no fractional bits. If the disparity is 16-bit signed format, as computed by @ref StereoBM or

  • q: Evision.Mat.t().

    \f$4 \times 4\f$ perspective transformation matrix that can be obtained with

Keyword Arguments
  • handleMissingValues: bool.

    Indicates, whether the function should handle missing values (i.e. points where the disparity was not computed). If handleMissingValues=true, then pixels with the minimal disparity that corresponds to the outliers (see StereoMatcher::compute ) are transformed to 3D points with a very large Z value (currently set to 10000).

  • ddepth: int.

    The optional output array depth. If it is -1, the output image will have CV_32F depth. ddepth can also be set to CV_16S, CV_32S or CV_32F.

Return
  • 3dImage: Evision.Mat.t().

    Output 3-channel floating-point image of the same size as disparity. Each element of _3dImage(x,y) contains 3D coordinates of the point (x,y) computed from the disparity map. If one uses Q obtained by @ref stereoRectify, then the returned points are represented in the first camera's rectified coordinate system.

@ref StereoSGBM and maybe other algorithms, it should be divided by 16 (and scaled to float) before being used here. @ref stereoRectify.

The function transforms a single-channel disparity map to a 3-channel image representing a 3D surface. That is, for each pixel (x,y) and the corresponding disparity d=disparity(x,y) , it computes: \f[\begin{bmatrix} X \\ Y \\ Z \\ W \end{bmatrix} = Q \begin{bmatrix} x \\ y \\ \texttt{disparity} (x,y) \\ z \end{bmatrix}.\f] @sa To reproject a sparse set of points {(x,y,d),...} to 3D space, use perspectiveTransform.

Python prototype (for reference only):

reprojectImageTo3D(disparity, Q[, _3dImage[, handleMissingValues[, ddepth]]]) -> _3dImage
@spec resize(
  Evision.Mat.maybe_mat_in(),
  {number(), number()}
) :: Evision.Mat.t() | {:error, String.t()}

Resizes an image.

Positional Arguments
  • src: Evision.Mat.t().

    input image.

  • dsize: Size.

    output image size; if it equals zero (None in Python), it is computed as: \f[\texttt{dsize = Size(round(fxsrc.cols), round(fysrc.rows))}\f] Either dsize or both fx and fy must be non-zero.

Keyword Arguments
  • fx: double.

    scale factor along the horizontal axis; when it equals 0, it is computed as \f[\texttt{(double)dsize.width/src.cols}\f]

  • fy: double.

    scale factor along the vertical axis; when it equals 0, it is computed as \f[\texttt{(double)dsize.height/src.rows}\f]

  • interpolation: int.

    interpolation method, see #InterpolationFlags

Return
  • dst: Evision.Mat.t().

    output image; it has the size dsize (when it is non-zero) or the size computed from src.size(), fx, and fy; the type of dst is the same as of src.

The function resize resizes the image src down to or up to the specified size. Note that the initial dst type or size are not taken into account. Instead, the size and type are derived from the src,dsize,fx, and fy. If you want to resize src so that it fits the pre-created dst, you may call the function as follows:

// explicitly specify dsize=dst.size(); fx and fy will be computed from that.
resize(src, dst, dst.size(), 0, 0, interpolation);

If you want to decimate the image by factor of 2 in each direction, you can call the function this way:

// specify fx and fy and let the function compute the destination image size.
resize(src, dst, Size(), 0.5, 0.5, interpolation);

To shrink an image, it will generally look best with #INTER_AREA interpolation, whereas to enlarge an image, it will generally look best with #INTER_CUBIC (slow) or #INTER_LINEAR (faster but still looks OK).

@sa warpAffine, warpPerspective, remap

Python prototype (for reference only):

resize(src, dsize[, dst[, fx[, fy[, interpolation]]]]) -> dst
Link to this function

resize(src, dsize, opts)

View Source
@spec resize(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Resizes an image.

Positional Arguments
  • src: Evision.Mat.t().

    input image.

  • dsize: Size.

    output image size; if it equals zero (None in Python), it is computed as: \f[\texttt{dsize = Size(round(fxsrc.cols), round(fysrc.rows))}\f] Either dsize or both fx and fy must be non-zero.

Keyword Arguments
  • fx: double.

    scale factor along the horizontal axis; when it equals 0, it is computed as \f[\texttt{(double)dsize.width/src.cols}\f]

  • fy: double.

    scale factor along the vertical axis; when it equals 0, it is computed as \f[\texttt{(double)dsize.height/src.rows}\f]

  • interpolation: int.

    interpolation method, see #InterpolationFlags

Return
  • dst: Evision.Mat.t().

    output image; it has the size dsize (when it is non-zero) or the size computed from src.size(), fx, and fy; the type of dst is the same as of src.

The function resize resizes the image src down to or up to the specified size. Note that the initial dst type or size are not taken into account. Instead, the size and type are derived from the src,dsize,fx, and fy. If you want to resize src so that it fits the pre-created dst, you may call the function as follows:

// explicitly specify dsize=dst.size(); fx and fy will be computed from that.
resize(src, dst, dst.size(), 0, 0, interpolation);

If you want to decimate the image by factor of 2 in each direction, you can call the function this way:

// specify fx and fy and let the function compute the destination image size.
resize(src, dst, Size(), 0.5, 0.5, interpolation);

To shrink an image, it will generally look best with #INTER_AREA interpolation, whereas to enlarge an image, it will generally look best with #INTER_CUBIC (slow) or #INTER_LINEAR (faster but still looks OK).

@sa warpAffine, warpPerspective, remap

Python prototype (for reference only):

resize(src, dsize[, dst[, fx[, fy[, interpolation]]]]) -> dst
Link to this function

resizeWindow(winname, size)

View Source
@spec resizeWindow(
  binary(),
  {number(), number()}
) :: :ok | {:error, String.t()}

resizeWindow

Positional Arguments
  • winname: String.

    Window name.

  • size: Size.

    The new window size.

Has overloading in C++

Python prototype (for reference only):

resizeWindow(winname, size) -> None
Link to this function

resizeWindow(winname, width, height)

View Source
@spec resizeWindow(binary(), integer(), integer()) :: :ok | {:error, String.t()}

Resizes the window to the specified size

Positional Arguments
  • winname: String.

    Window name.

  • width: int.

    The new window width.

  • height: int.

    The new window height.

Note: The specified window size is for the image area. Toolbars are not counted. Only windows created without cv::WINDOW_AUTOSIZE flag can be resized.

Python prototype (for reference only):

resizeWindow(winname, width, height) -> None
@spec rodrigues(Evision.Mat.maybe_mat_in()) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Converts a rotation matrix to a rotation vector or vice versa.

Positional Arguments
  • src: Evision.Mat.t().

    Input rotation vector (3x1 or 1x3) or rotation matrix (3x3).

Return
  • dst: Evision.Mat.t().

    Output rotation matrix (3x3) or rotation vector (3x1 or 1x3), respectively.

  • jacobian: Evision.Mat.t().

    Optional output Jacobian matrix, 3x9 or 9x3, which is a matrix of partial derivatives of the output array components with respect to the input array components.

\f[\begin{array}{l} \theta \leftarrow norm(r) \\ r \leftarrow r/ \theta \\ R = \cos(\theta) I + (1- \cos{\theta} ) r r^T + \sin(\theta) \vecthreethree{0}{-r_z}{r_y}{r_z}{0}{-r_x}{-r_y}{r_x}{0} \end{array}\f] Inverse transformation can be also done easily, since \f[\sin ( \theta ) \vecthreethree{0}{-r_z}{r_y}{r_z}{0}{-r_x}{-r_y}{r_x}{0} = \frac{R - R^T}{2}\f] A rotation vector is a convenient and most compact representation of a rotation matrix (since any rotation matrix has just 3 degrees of freedom). The representation is used in the global 3D geometry optimization procedures like @ref calibrateCamera, @ref stereoCalibrate, or @ref solvePnP . Note: More information about the computation of the derivative of a 3D rotation matrix with respect to its exponential coordinate can be found in:

  • A Compact Formula for the Derivative of a 3-D Rotation in Exponential Coordinates, Guillermo Gallego, Anthony J. Yezzi @cite Gallego2014ACF

Note: Useful information on SE(3) and Lie Groups can be found in:

  • A tutorial on SE(3) transformation parameterizations and on-manifold optimization, Jose-Luis Blanco @cite blanco2010tutorial
  • Lie Groups for 2D and 3D Transformation, Ethan Eade @cite Eade17
  • A micro Lie theory for state estimation in robotics, Joan Solà, Jérémie Deray, Dinesh Atchuthan @cite Sol2018AML

Python prototype (for reference only):

Rodrigues(src[, dst[, jacobian]]) -> dst, jacobian
@spec rodrigues(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Converts a rotation matrix to a rotation vector or vice versa.

Positional Arguments
  • src: Evision.Mat.t().

    Input rotation vector (3x1 or 1x3) or rotation matrix (3x3).

Return
  • dst: Evision.Mat.t().

    Output rotation matrix (3x3) or rotation vector (3x1 or 1x3), respectively.

  • jacobian: Evision.Mat.t().

    Optional output Jacobian matrix, 3x9 or 9x3, which is a matrix of partial derivatives of the output array components with respect to the input array components.

\f[\begin{array}{l} \theta \leftarrow norm(r) \\ r \leftarrow r/ \theta \\ R = \cos(\theta) I + (1- \cos{\theta} ) r r^T + \sin(\theta) \vecthreethree{0}{-r_z}{r_y}{r_z}{0}{-r_x}{-r_y}{r_x}{0} \end{array}\f] Inverse transformation can be also done easily, since \f[\sin ( \theta ) \vecthreethree{0}{-r_z}{r_y}{r_z}{0}{-r_x}{-r_y}{r_x}{0} = \frac{R - R^T}{2}\f] A rotation vector is a convenient and most compact representation of a rotation matrix (since any rotation matrix has just 3 degrees of freedom). The representation is used in the global 3D geometry optimization procedures like @ref calibrateCamera, @ref stereoCalibrate, or @ref solvePnP . Note: More information about the computation of the derivative of a 3D rotation matrix with respect to its exponential coordinate can be found in:

  • A Compact Formula for the Derivative of a 3-D Rotation in Exponential Coordinates, Guillermo Gallego, Anthony J. Yezzi @cite Gallego2014ACF

Note: Useful information on SE(3) and Lie Groups can be found in:

  • A tutorial on SE(3) transformation parameterizations and on-manifold optimization, Jose-Luis Blanco @cite blanco2010tutorial
  • Lie Groups for 2D and 3D Transformation, Ethan Eade @cite Eade17
  • A micro Lie theory for state estimation in robotics, Joan Solà, Jérémie Deray, Dinesh Atchuthan @cite Sol2018AML

Python prototype (for reference only):

Rodrigues(src[, dst[, jacobian]]) -> dst, jacobian
@spec rotate(Evision.Mat.maybe_mat_in(), integer()) ::
  Evision.Mat.t() | {:error, String.t()}

Rotates a 2D array in multiples of 90 degrees. The function cv::rotate rotates the array in one of three different ways: Rotate by 90 degrees clockwise (rotateCode = ROTATE_90_CLOCKWISE). Rotate by 180 degrees clockwise (rotateCode = ROTATE_180). Rotate by 270 degrees clockwise (rotateCode = ROTATE_90_COUNTERCLOCKWISE).

Positional Arguments
  • src: Evision.Mat.t().

    input array.

  • rotateCode: int.

    an enum to specify how to rotate the array; see the enum #RotateFlags

Return
  • dst: Evision.Mat.t().

    output array of the same type as src. The size is the same with ROTATE_180, and the rows and cols are switched for ROTATE_90_CLOCKWISE and ROTATE_90_COUNTERCLOCKWISE.

@sa transpose , repeat , completeSymm, flip, RotateFlags

Python prototype (for reference only):

rotate(src, rotateCode[, dst]) -> dst
Link to this function

rotate(src, rotateCode, opts)

View Source
@spec rotate(Evision.Mat.maybe_mat_in(), integer(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Rotates a 2D array in multiples of 90 degrees. The function cv::rotate rotates the array in one of three different ways: Rotate by 90 degrees clockwise (rotateCode = ROTATE_90_CLOCKWISE). Rotate by 180 degrees clockwise (rotateCode = ROTATE_180). Rotate by 270 degrees clockwise (rotateCode = ROTATE_90_COUNTERCLOCKWISE).

Positional Arguments
  • src: Evision.Mat.t().

    input array.

  • rotateCode: int.

    an enum to specify how to rotate the array; see the enum #RotateFlags

Return
  • dst: Evision.Mat.t().

    output array of the same type as src. The size is the same with ROTATE_180, and the rows and cols are switched for ROTATE_90_CLOCKWISE and ROTATE_90_COUNTERCLOCKWISE.

@sa transpose , repeat , completeSymm, flip, RotateFlags

Python prototype (for reference only):

rotate(src, rotateCode[, dst]) -> dst
Link to this function

rotatedRectangleIntersection(rect1, rect2)

View Source
@spec rotatedRectangleIntersection(Evision.RotatedRect.t(), Evision.RotatedRect.t()) ::
  {integer(), Evision.Mat.t()} | {:error, String.t()}

Finds out if there is any intersection between two rotated rectangles.

Positional Arguments
  • rect1: {centre={x, y}, size={s1, s2}, angle}.

    First rectangle

  • rect2: {centre={x, y}, size={s1, s2}, angle}.

    Second rectangle

Return
  • retval: int

  • intersectingRegion: Evision.Mat.t().

    The output array of the vertices of the intersecting region. It returns at most 8 vertices. Stored as std::vector\<cv::Point2f> or cv::Mat as Mx1 of type CV_32FC2.

If there is then the vertices of the intersecting region are returned as well. Below are some examples of intersection configurations. The hatched pattern indicates the intersecting region and the red vertices are returned by the function. intersection examples @returns One of #RectanglesIntersectTypes

Python prototype (for reference only):

rotatedRectangleIntersection(rect1, rect2[, intersectingRegion]) -> retval, intersectingRegion
Link to this function

rotatedRectangleIntersection(rect1, rect2, opts)

View Source
@spec rotatedRectangleIntersection(
  Evision.RotatedRect.t(),
  Evision.RotatedRect.t(),
  [{atom(), term()}, ...] | nil
) :: {integer(), Evision.Mat.t()} | {:error, String.t()}

Finds out if there is any intersection between two rotated rectangles.

Positional Arguments
  • rect1: {centre={x, y}, size={s1, s2}, angle}.

    First rectangle

  • rect2: {centre={x, y}, size={s1, s2}, angle}.

    Second rectangle

Return
  • retval: int

  • intersectingRegion: Evision.Mat.t().

    The output array of the vertices of the intersecting region. It returns at most 8 vertices. Stored as std::vector\<cv::Point2f> or cv::Mat as Mx1 of type CV_32FC2.

If there is then the vertices of the intersecting region are returned as well. Below are some examples of intersection configurations. The hatched pattern indicates the intersecting region and the red vertices are returned by the function. intersection examples @returns One of #RectanglesIntersectTypes

Python prototype (for reference only):

rotatedRectangleIntersection(rect1, rect2[, intersectingRegion]) -> retval, intersectingRegion

Computes an RQ decomposition of 3x3 matrices.

Positional Arguments
  • src: Evision.Mat.t().

    3x3 input matrix.

Return
  • retval: Vec3d

  • mtxR: Evision.Mat.t().

    Output 3x3 upper-triangular matrix.

  • mtxQ: Evision.Mat.t().

    Output 3x3 orthogonal matrix.

  • qx: Evision.Mat.t().

    Optional output 3x3 rotation matrix around x-axis.

  • qy: Evision.Mat.t().

    Optional output 3x3 rotation matrix around y-axis.

  • qz: Evision.Mat.t().

    Optional output 3x3 rotation matrix around z-axis.

The function computes a RQ decomposition using the given rotations. This function is used in #decomposeProjectionMatrix to decompose the left 3x3 submatrix of a projection matrix into a camera and a rotation matrix. It optionally returns three rotation matrices, one for each axis, and the three Euler angles in degrees (as the return value) that could be used in OpenGL. Note, there is always more than one sequence of rotations about the three principal axes that results in the same orientation of an object, e.g. see @cite Slabaugh . Returned three rotation matrices and corresponding three Euler angles are only one of the possible solutions.

Python prototype (for reference only):

RQDecomp3x3(src[, mtxR[, mtxQ[, Qx[, Qy[, Qz]]]]]) -> retval, mtxR, mtxQ, Qx, Qy, Qz
@spec rqDecomp3x3(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  {{number(), number(), number()}, Evision.Mat.t(), Evision.Mat.t(),
   Evision.Mat.t(), Evision.Mat.t(), Evision.Mat.t()}
  | {:error, String.t()}

Computes an RQ decomposition of 3x3 matrices.

Positional Arguments
  • src: Evision.Mat.t().

    3x3 input matrix.

Return
  • retval: Vec3d

  • mtxR: Evision.Mat.t().

    Output 3x3 upper-triangular matrix.

  • mtxQ: Evision.Mat.t().

    Output 3x3 orthogonal matrix.

  • qx: Evision.Mat.t().

    Optional output 3x3 rotation matrix around x-axis.

  • qy: Evision.Mat.t().

    Optional output 3x3 rotation matrix around y-axis.

  • qz: Evision.Mat.t().

    Optional output 3x3 rotation matrix around z-axis.

The function computes a RQ decomposition using the given rotations. This function is used in #decomposeProjectionMatrix to decompose the left 3x3 submatrix of a projection matrix into a camera and a rotation matrix. It optionally returns three rotation matrices, one for each axis, and the three Euler angles in degrees (as the return value) that could be used in OpenGL. Note, there is always more than one sequence of rotations about the three principal axes that results in the same orientation of an object, e.g. see @cite Slabaugh . Returned three rotation matrices and corresponding three Euler angles are only one of the possible solutions.

Python prototype (for reference only):

RQDecomp3x3(src[, mtxR[, mtxQ[, Qx[, Qy[, Qz]]]]]) -> retval, mtxR, mtxQ, Qx, Qy, Qz
Link to this function

sampsonDistance(pt1, pt2, f)

View Source
@spec sampsonDistance(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in()
) :: number() | {:error, String.t()}

Calculates the Sampson Distance between two points.

Positional Arguments
  • pt1: Evision.Mat.t().

    first homogeneous 2d point

  • pt2: Evision.Mat.t().

    second homogeneous 2d point

  • f: Evision.Mat.t().

    fundamental matrix

Return
  • retval: double

The function cv::sampsonDistance calculates and returns the first order approximation of the geometric error as: \f[ sd( \texttt{pt1} , \texttt{pt2} )= \frac{(\texttt{pt2}^t \cdot \texttt{F} \cdot \texttt{pt1})^2} {((\texttt{F} \cdot \texttt{pt1})(0))^2 + ((\texttt{F} \cdot \texttt{pt1})(1))^2 + ((\texttt{F}^t \cdot \texttt{pt2})(0))^2 + ((\texttt{F}^t \cdot \texttt{pt2})(1))^2} \f] The fundamental matrix may be calculated using the #findFundamentalMat function. See @cite HartleyZ00 11.4.3 for details. @return The computed Sampson distance.

Python prototype (for reference only):

sampsonDistance(pt1, pt2, F) -> retval
Link to this function

scaleAdd(src1, alpha, src2)

View Source

Calculates the sum of a scaled array and another array.

Positional Arguments
  • src1: Evision.Mat.t().

    first input array.

  • alpha: double.

    scale factor for the first array.

  • src2: Evision.Mat.t().

    second input array of the same size and type as src1.

Return
  • dst: Evision.Mat.t().

    output array of the same size and type as src1.

The function scaleAdd is one of the classical primitive linear algebra operations, known as DAXPY or SAXPY in BLAS. It calculates the sum of a scaled array and another array: \f[\texttt{dst} (I)= \texttt{scale} \cdot \texttt{src1} (I) + \texttt{src2} (I)\f] The function can also be emulated with a matrix expression, for example:

Mat A(3, 3, CV_64F);
...
A.row(0) = A.row(1)*2 + A.row(2);

@sa add, addWeighted, subtract, Mat::dot, Mat::convertTo

Python prototype (for reference only):

scaleAdd(src1, alpha, src2[, dst]) -> dst
Link to this function

scaleAdd(src1, alpha, src2, opts)

View Source
@spec scaleAdd(
  Evision.Mat.maybe_mat_in(),
  number(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Calculates the sum of a scaled array and another array.

Positional Arguments
  • src1: Evision.Mat.t().

    first input array.

  • alpha: double.

    scale factor for the first array.

  • src2: Evision.Mat.t().

    second input array of the same size and type as src1.

Return
  • dst: Evision.Mat.t().

    output array of the same size and type as src1.

The function scaleAdd is one of the classical primitive linear algebra operations, known as DAXPY or SAXPY in BLAS. It calculates the sum of a scaled array and another array: \f[\texttt{dst} (I)= \texttt{scale} \cdot \texttt{src1} (I) + \texttt{src2} (I)\f] The function can also be emulated with a matrix expression, for example:

Mat A(3, 3, CV_64F);
...
A.row(0) = A.row(1)*2 + A.row(2);

@sa add, addWeighted, subtract, Mat::dot, Mat::convertTo

Python prototype (for reference only):

scaleAdd(src1, alpha, src2[, dst]) -> dst
Link to this function

scharr(src, ddepth, dx, dy)

View Source
@spec scharr(Evision.Mat.maybe_mat_in(), integer(), integer(), integer()) ::
  Evision.Mat.t() | {:error, String.t()}

Calculates the first x- or y- image derivative using Scharr operator.

Positional Arguments
  • src: Evision.Mat.t().

    input image.

  • ddepth: int.

    output image depth, see @ref filter_depths "combinations"

  • dx: int.

    order of the derivative x.

  • dy: int.

    order of the derivative y.

Keyword Arguments
  • scale: double.

    optional scale factor for the computed derivative values; by default, no scaling is applied (see #getDerivKernels for details).

  • delta: double.

    optional delta value that is added to the results prior to storing them in dst.

  • borderType: int.

    pixel extrapolation method, see #BorderTypes. #BORDER_WRAP is not supported.

Return
  • dst: Evision.Mat.t().

    output image of the same size and the same number of channels as src.

The function computes the first x- or y- spatial image derivative using the Scharr operator. The call \f[\texttt{Scharr(src, dst, ddepth, dx, dy, scale, delta, borderType)}\f] is equivalent to \f[\texttt{Sobel(src, dst, ddepth, dx, dy, FILTER_SCHARR, scale, delta, borderType)} .\f] @sa cartToPolar

Python prototype (for reference only):

Scharr(src, ddepth, dx, dy[, dst[, scale[, delta[, borderType]]]]) -> dst
Link to this function

scharr(src, ddepth, dx, dy, opts)

View Source
@spec scharr(
  Evision.Mat.maybe_mat_in(),
  integer(),
  integer(),
  integer(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Calculates the first x- or y- image derivative using Scharr operator.

Positional Arguments
  • src: Evision.Mat.t().

    input image.

  • ddepth: int.

    output image depth, see @ref filter_depths "combinations"

  • dx: int.

    order of the derivative x.

  • dy: int.

    order of the derivative y.

Keyword Arguments
  • scale: double.

    optional scale factor for the computed derivative values; by default, no scaling is applied (see #getDerivKernels for details).

  • delta: double.

    optional delta value that is added to the results prior to storing them in dst.

  • borderType: int.

    pixel extrapolation method, see #BorderTypes. #BORDER_WRAP is not supported.

Return
  • dst: Evision.Mat.t().

    output image of the same size and the same number of channels as src.

The function computes the first x- or y- spatial image derivative using the Scharr operator. The call \f[\texttt{Scharr(src, dst, ddepth, dx, dy, scale, delta, borderType)}\f] is equivalent to \f[\texttt{Sobel(src, dst, ddepth, dx, dy, FILTER_SCHARR, scale, delta, borderType)} .\f] @sa cartToPolar

Python prototype (for reference only):

Scharr(src, ddepth, dx, dy[, dst[, scale[, delta[, borderType]]]]) -> dst
Link to this function

seamlessClone(src, dst, mask, p, flags)

View Source

Image editing tasks concern either global changes (color/intensity corrections, filters, deformations) or local changes concerned to a selection. Here we are interested in achieving local changes, ones that are restricted to a region manually selected (ROI), in a seamless and effortless manner. The extent of the changes ranges from slight distortions to complete replacement by novel content @cite PM03 .

Positional Arguments
  • src: Evision.Mat.t().

    Input 8-bit 3-channel image.

  • dst: Evision.Mat.t().

    Input 8-bit 3-channel image.

  • mask: Evision.Mat.t().

    Input 8-bit 1 or 3-channel image.

  • p: Point.

    Point in dst image where object is placed.

  • flags: int.

    Cloning method that could be cv::NORMAL_CLONE, cv::MIXED_CLONE or cv::MONOCHROME_TRANSFER

Return
  • blend: Evision.Mat.t().

    Output image with the same size and type as dst.

Python prototype (for reference only):

seamlessClone(src, dst, mask, p, flags[, blend]) -> blend
Link to this function

seamlessClone(src, dst, mask, p, flags, opts)

View Source
@spec seamlessClone(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  integer(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Image editing tasks concern either global changes (color/intensity corrections, filters, deformations) or local changes concerned to a selection. Here we are interested in achieving local changes, ones that are restricted to a region manually selected (ROI), in a seamless and effortless manner. The extent of the changes ranges from slight distortions to complete replacement by novel content @cite PM03 .

Positional Arguments
  • src: Evision.Mat.t().

    Input 8-bit 3-channel image.

  • dst: Evision.Mat.t().

    Input 8-bit 3-channel image.

  • mask: Evision.Mat.t().

    Input 8-bit 1 or 3-channel image.

  • p: Point.

    Point in dst image where object is placed.

  • flags: int.

    Cloning method that could be cv::NORMAL_CLONE, cv::MIXED_CLONE or cv::MONOCHROME_TRANSFER

Return
  • blend: Evision.Mat.t().

    Output image with the same size and type as dst.

Python prototype (for reference only):

seamlessClone(src, dst, mask, p, flags[, blend]) -> blend
@spec selectROI(Evision.Mat.maybe_mat_in()) ::
  {number(), number(), number(), number()} | {:error, String.t()}

selectROI

Positional Arguments
  • img: Evision.Mat.t()
Keyword Arguments
  • showCrosshair: bool.
  • fromCenter: bool.
  • printNotice: bool.
Return
  • retval: Rect

Has overloading in C++

Python prototype (for reference only):

selectROI(img[, showCrosshair[, fromCenter[, printNotice]]]) -> retval
@spec selectROI(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  {number(), number(), number(), number()} | {:error, String.t()}
@spec selectROI(binary(), Evision.Mat.maybe_mat_in()) ::
  {number(), number(), number(), number()} | {:error, String.t()}

Variant 1:

Allows users to select a ROI on the given image.

Positional Arguments
  • windowName: String.

    name of the window where selection process will be shown.

  • img: Evision.Mat.t().

    image to select a ROI.

Keyword Arguments
  • showCrosshair: bool.

    if true crosshair of selection rectangle will be shown.

  • fromCenter: bool.

    if true center of selection will match initial mouse position. In opposite case a corner of selection rectangle will correspont to the initial mouse position.

  • printNotice: bool.

    if true a notice to select ROI or cancel selection will be printed in console.

Return
  • retval: Rect

The function creates a window and allows users to select a ROI using the mouse. Controls: use space or enter to finish selection, use key c to cancel selection (function will return the zero cv::Rect). @return selected ROI or empty rect if selection canceled. Note: The function sets it's own mouse callback for specified window using cv::setMouseCallback(windowName, ...). After finish of work an empty callback will be set for the used window.

Python prototype (for reference only):

selectROI(windowName, img[, showCrosshair[, fromCenter[, printNotice]]]) -> retval

Variant 2:

selectROI

Positional Arguments
  • img: Evision.Mat.t()
Keyword Arguments
  • showCrosshair: bool.
  • fromCenter: bool.
  • printNotice: bool.
Return
  • retval: Rect

Has overloading in C++

Python prototype (for reference only):

selectROI(img[, showCrosshair[, fromCenter[, printNotice]]]) -> retval
Link to this function

selectROI(windowName, img, opts)

View Source
@spec selectROI(binary(), Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  {number(), number(), number(), number()} | {:error, String.t()}

Allows users to select a ROI on the given image.

Positional Arguments
  • windowName: String.

    name of the window where selection process will be shown.

  • img: Evision.Mat.t().

    image to select a ROI.

Keyword Arguments
  • showCrosshair: bool.

    if true crosshair of selection rectangle will be shown.

  • fromCenter: bool.

    if true center of selection will match initial mouse position. In opposite case a corner of selection rectangle will correspont to the initial mouse position.

  • printNotice: bool.

    if true a notice to select ROI or cancel selection will be printed in console.

Return
  • retval: Rect

The function creates a window and allows users to select a ROI using the mouse. Controls: use space or enter to finish selection, use key c to cancel selection (function will return the zero cv::Rect). @return selected ROI or empty rect if selection canceled. Note: The function sets it's own mouse callback for specified window using cv::setMouseCallback(windowName, ...). After finish of work an empty callback will be set for the used window.

Python prototype (for reference only):

selectROI(windowName, img[, showCrosshair[, fromCenter[, printNotice]]]) -> retval
Link to this function

selectROIs(windowName, img)

View Source
@spec selectROIs(binary(), Evision.Mat.maybe_mat_in()) ::
  [{number(), number(), number(), number()}] | {:error, String.t()}

Allows users to select multiple ROIs on the given image.

Positional Arguments
  • windowName: String.

    name of the window where selection process will be shown.

  • img: Evision.Mat.t().

    image to select a ROI.

Keyword Arguments
  • showCrosshair: bool.

    if true crosshair of selection rectangle will be shown.

  • fromCenter: bool.

    if true center of selection will match initial mouse position. In opposite case a corner of selection rectangle will correspont to the initial mouse position.

  • printNotice: bool.

    if true a notice to select ROI or cancel selection will be printed in console.

Return
  • boundingBoxes: [Rect].

    selected ROIs.

The function creates a window and allows users to select multiple ROIs using the mouse. Controls: use space or enter to finish current selection and start a new one, use esc to terminate multiple ROI selection process.

Note: The function sets it's own mouse callback for specified window using cv::setMouseCallback(windowName, ...). After finish of work an empty callback will be set for the used window.

Python prototype (for reference only):

selectROIs(windowName, img[, showCrosshair[, fromCenter[, printNotice]]]) -> boundingBoxes
Link to this function

selectROIs(windowName, img, opts)

View Source
@spec selectROIs(binary(), Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  [{number(), number(), number(), number()}] | {:error, String.t()}

Allows users to select multiple ROIs on the given image.

Positional Arguments
  • windowName: String.

    name of the window where selection process will be shown.

  • img: Evision.Mat.t().

    image to select a ROI.

Keyword Arguments
  • showCrosshair: bool.

    if true crosshair of selection rectangle will be shown.

  • fromCenter: bool.

    if true center of selection will match initial mouse position. In opposite case a corner of selection rectangle will correspont to the initial mouse position.

  • printNotice: bool.

    if true a notice to select ROI or cancel selection will be printed in console.

Return
  • boundingBoxes: [Rect].

    selected ROIs.

The function creates a window and allows users to select multiple ROIs using the mouse. Controls: use space or enter to finish current selection and start a new one, use esc to terminate multiple ROI selection process.

Note: The function sets it's own mouse callback for specified window using cv::setMouseCallback(windowName, ...). After finish of work an empty callback will be set for the used window.

Python prototype (for reference only):

selectROIs(windowName, img[, showCrosshair[, fromCenter[, printNotice]]]) -> boundingBoxes
Link to this function

sepFilter2D(src, ddepth, kernelX, kernelY)

View Source

Applies a separable linear filter to an image.

Positional Arguments
  • src: Evision.Mat.t().

    Source image.

  • ddepth: int.

    Destination image depth, see @ref filter_depths "combinations"

  • kernelX: Evision.Mat.t().

    Coefficients for filtering each row.

  • kernelY: Evision.Mat.t().

    Coefficients for filtering each column.

Keyword Arguments
  • anchor: Point.

    Anchor position within the kernel. The default value \f$(-1,-1)\f$ means that the anchor is at the kernel center.

  • delta: double.

    Value added to the filtered results before storing them.

  • borderType: int.

    Pixel extrapolation method, see #BorderTypes. #BORDER_WRAP is not supported.

Return
  • dst: Evision.Mat.t().

    Destination image of the same size and the same number of channels as src .

The function applies a separable linear filter to the image. That is, first, every row of src is filtered with the 1D kernel kernelX. Then, every column of the result is filtered with the 1D kernel kernelY. The final result shifted by delta is stored in dst . @sa filter2D, Sobel, GaussianBlur, boxFilter, blur

Python prototype (for reference only):

sepFilter2D(src, ddepth, kernelX, kernelY[, dst[, anchor[, delta[, borderType]]]]) -> dst
Link to this function

sepFilter2D(src, ddepth, kernelX, kernelY, opts)

View Source
@spec sepFilter2D(
  Evision.Mat.maybe_mat_in(),
  integer(),
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Applies a separable linear filter to an image.

Positional Arguments
  • src: Evision.Mat.t().

    Source image.

  • ddepth: int.

    Destination image depth, see @ref filter_depths "combinations"

  • kernelX: Evision.Mat.t().

    Coefficients for filtering each row.

  • kernelY: Evision.Mat.t().

    Coefficients for filtering each column.

Keyword Arguments
  • anchor: Point.

    Anchor position within the kernel. The default value \f$(-1,-1)\f$ means that the anchor is at the kernel center.

  • delta: double.

    Value added to the filtered results before storing them.

  • borderType: int.

    Pixel extrapolation method, see #BorderTypes. #BORDER_WRAP is not supported.

Return
  • dst: Evision.Mat.t().

    Destination image of the same size and the same number of channels as src .

The function applies a separable linear filter to the image. That is, first, every row of src is filtered with the 1D kernel kernelX. Then, every column of the result is filtered with the 1D kernel kernelY. The final result shifted by delta is stored in dst . @sa filter2D, Sobel, GaussianBlur, boxFilter, blur

Python prototype (for reference only):

sepFilter2D(src, ddepth, kernelX, kernelY[, dst[, anchor[, delta[, borderType]]]]) -> dst
@spec setIdentity(Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

Initializes a scaled identity matrix.

Keyword Arguments
  • s: Scalar.

    value to assign to diagonal elements.

Return
  • mtx: Evision.Mat.t().

    matrix to initialize (not necessarily square).

The function cv::setIdentity initializes a scaled identity matrix: \f[\texttt{mtx} (i,j)= \fork{\texttt{value}}{ if (i=j)}{0}{otherwise}\f] The function can also be emulated using the matrix initializers and the matrix expressions:

Mat A = Mat::eye(4, 3, CV_32F)*5;
// A will be set to [[5, 0, 0], [0, 5, 0], [0, 0, 5], [0, 0, 0]]

@sa Mat::zeros, Mat::ones, Mat::setTo, Mat::operator=

Python prototype (for reference only):

setIdentity(mtx[, s]) -> mtx
@spec setIdentity(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Initializes a scaled identity matrix.

Keyword Arguments
  • s: Scalar.

    value to assign to diagonal elements.

Return
  • mtx: Evision.Mat.t().

    matrix to initialize (not necessarily square).

The function cv::setIdentity initializes a scaled identity matrix: \f[\texttt{mtx} (i,j)= \fork{\texttt{value}}{ if (i=j)}{0}{otherwise}\f] The function can also be emulated using the matrix initializers and the matrix expressions:

Mat A = Mat::eye(4, 3, CV_32F)*5;
// A will be set to [[5, 0, 0], [0, 5, 0], [0, 0, 5], [0, 0, 0]]

@sa Mat::zeros, Mat::ones, Mat::setTo, Mat::operator=

Python prototype (for reference only):

setIdentity(mtx[, s]) -> mtx
@spec setLogLevel(integer()) :: integer() | {:error, String.t()}

setLogLevel

Positional Arguments
  • level: int
Return
  • retval: int

Python prototype (for reference only):

setLogLevel(level) -> retval
@spec setNumThreads(integer()) :: :ok | {:error, String.t()}

OpenCV will try to set the number of threads for subsequent parallel regions.

Positional Arguments
  • nthreads: int.

    Number of threads used by OpenCV.

If threads == 1, OpenCV will disable threading optimizations and run all it's functions sequentially. Passing threads \< 0 will reset threads number to system default. The function is not thread-safe. It must not be called in parallel region or concurrent threads. OpenCV will try to run its functions with specified threads number, but some behaviour differs from framework:

  • TBB - User-defined parallel constructions will run with the same threads number, if another is not specified. If later on user creates his own scheduler, OpenCV will use it.

  • OpenMP - No special defined behaviour.

  • Concurrency - If threads == 1, OpenCV will disable threading optimizations and run its functions sequentially.

  • GCD - Supports only values \<= 0.

  • C= - No special defined behaviour. @sa getNumThreads, getThreadNum

Python prototype (for reference only):

setNumThreads(nthreads) -> None
@spec setRNGSeed(integer()) :: :ok | {:error, String.t()}

Sets state of default random number generator.

Positional Arguments
  • seed: int.

    new state for default random number generator

The function cv::setRNGSeed sets state of default random number generator to custom value. @sa RNG, randu, randn

Python prototype (for reference only):

setRNGSeed(seed) -> None
Link to this function

setTrackbarMax(trackbarname, winname, maxval)

View Source
@spec setTrackbarMax(binary(), binary(), integer()) :: :ok | {:error, String.t()}

Sets the trackbar maximum position.

Positional Arguments
  • trackbarname: String.

    Name of the trackbar.

  • winname: String.

    Name of the window that is the parent of trackbar.

  • maxval: int.

    New maximum position.

The function sets the maximum position of the specified trackbar in the specified window. Note: [Qt Backend Only] winname can be empty if the trackbar is attached to the control panel.

Python prototype (for reference only):

setTrackbarMax(trackbarname, winname, maxval) -> None
Link to this function

setTrackbarMin(trackbarname, winname, minval)

View Source
@spec setTrackbarMin(binary(), binary(), integer()) :: :ok | {:error, String.t()}

Sets the trackbar minimum position.

Positional Arguments
  • trackbarname: String.

    Name of the trackbar.

  • winname: String.

    Name of the window that is the parent of trackbar.

  • minval: int.

    New minimum position.

The function sets the minimum position of the specified trackbar in the specified window. Note: [Qt Backend Only] winname can be empty if the trackbar is attached to the control panel.

Python prototype (for reference only):

setTrackbarMin(trackbarname, winname, minval) -> None
Link to this function

setTrackbarPos(trackbarname, winname, pos)

View Source
@spec setTrackbarPos(binary(), binary(), integer()) :: :ok | {:error, String.t()}

Sets the trackbar position.

Positional Arguments
  • trackbarname: String.

    Name of the trackbar.

  • winname: String.

    Name of the window that is the parent of trackbar.

  • pos: int.

    New position.

The function sets the position of the specified trackbar in the specified window. Note: [Qt Backend Only] winname can be empty if the trackbar is attached to the control panel.

Python prototype (for reference only):

setTrackbarPos(trackbarname, winname, pos) -> None
@spec setUseOpenVX(boolean()) :: :ok | {:error, String.t()}

setUseOpenVX

Positional Arguments
  • flag: bool

Python prototype (for reference only):

setUseOpenVX(flag) -> None
@spec setUseOptimized(boolean()) :: :ok | {:error, String.t()}

Enables or disables the optimized code.

Positional Arguments
  • onoff: bool.

    The boolean flag specifying whether the optimized code should be used (onoff=true) or not (onoff=false).

The function can be used to dynamically turn on and off optimized dispatched code (code that uses SSE4.2, AVX/AVX2, and other instructions on the platforms that support it). It sets a global flag that is further checked by OpenCV functions. Since the flag is not checked in the inner OpenCV loops, it is only safe to call the function on the very top level in your application where you can be sure that no other OpenCV function is currently executed. By default, the optimized code is enabled unless you disable it in CMake. The current status can be retrieved using useOptimized.

Python prototype (for reference only):

setUseOptimized(onoff) -> None
Link to this function

setWindowProperty(winname, prop_id, prop_value)

View Source
@spec setWindowProperty(binary(), integer(), number()) :: :ok | {:error, String.t()}

Changes parameters of a window dynamically.

Positional Arguments
  • winname: String.

    Name of the window.

  • prop_id: int.

    Window property to edit. The supported operation flags are: (cv::WindowPropertyFlags)

  • prop_value: double.

    New value of the window property. The supported flags are: (cv::WindowFlags)

The function setWindowProperty enables changing properties of a window.

Python prototype (for reference only):

setWindowProperty(winname, prop_id, prop_value) -> None
Link to this function

setWindowTitle(winname, title)

View Source
@spec setWindowTitle(binary(), binary()) :: :ok | {:error, String.t()}

Updates window title

Positional Arguments

Python prototype (for reference only):

setWindowTitle(winname, title) -> None
Link to this function

sobel(src, ddepth, dx, dy)

View Source
@spec sobel(Evision.Mat.maybe_mat_in(), integer(), integer(), integer()) ::
  Evision.Mat.t() | {:error, String.t()}

Calculates the first, second, third, or mixed image derivatives using an extended Sobel operator.

Positional Arguments
  • src: Evision.Mat.t().

    input image.

  • ddepth: int.

    output image depth, see @ref filter_depths "combinations"; in the case of 8-bit input images it will result in truncated derivatives.

  • dx: int.

    order of the derivative x.

  • dy: int.

    order of the derivative y.

Keyword Arguments
  • ksize: int.

    size of the extended Sobel kernel; it must be 1, 3, 5, or 7.

  • scale: double.

    optional scale factor for the computed derivative values; by default, no scaling is applied (see #getDerivKernels for details).

  • delta: double.

    optional delta value that is added to the results prior to storing them in dst.

  • borderType: int.

    pixel extrapolation method, see #BorderTypes. #BORDER_WRAP is not supported.

Return
  • dst: Evision.Mat.t().

    output image of the same size and the same number of channels as src .

In all cases except one, the \f$\texttt{ksize} \times \texttt{ksize}\f$ separable kernel is used to calculate the derivative. When \f$\texttt{ksize = 1}\f$, the \f$3 \times 1\f$ or \f$1 \times 3\f$ kernel is used (that is, no Gaussian smoothing is done). ksize = 1 can only be used for the first or the second x- or y- derivatives. There is also the special value ksize = #FILTER_SCHARR (-1) that corresponds to the \f$3\times3\f$ Scharr filter that may give more accurate results than the \f$3\times3\f$ Sobel. The Scharr aperture is \f[\vecthreethree{-3}{0}{3}{-10}{0}{10}{-3}{0}{3}\f] for the x-derivative, or transposed for the y-derivative. The function calculates an image derivative by convolving the image with the appropriate kernel: \f[\texttt{dst} = \frac{\partial^{xorder+yorder} \texttt{src}}{\partial x^{xorder} \partial y^{yorder}}\f] The Sobel operators combine Gaussian smoothing and differentiation, so the result is more or less resistant to the noise. Most often, the function is called with ( xorder = 1, yorder = 0, ksize = 3) or ( xorder = 0, yorder = 1, ksize = 3) to calculate the first x- or y- image derivative. The first case corresponds to a kernel of: \f[\vecthreethree{-1}{0}{1}{-2}{0}{2}{-1}{0}{1}\f] The second case corresponds to a kernel of: \f[\vecthreethree{-1}{-2}{-1}{0}{0}{0}{1}{2}{1}\f] @sa Scharr, Laplacian, sepFilter2D, filter2D, GaussianBlur, cartToPolar

Python prototype (for reference only):

Sobel(src, ddepth, dx, dy[, dst[, ksize[, scale[, delta[, borderType]]]]]) -> dst
Link to this function

sobel(src, ddepth, dx, dy, opts)

View Source
@spec sobel(
  Evision.Mat.maybe_mat_in(),
  integer(),
  integer(),
  integer(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Calculates the first, second, third, or mixed image derivatives using an extended Sobel operator.

Positional Arguments
  • src: Evision.Mat.t().

    input image.

  • ddepth: int.

    output image depth, see @ref filter_depths "combinations"; in the case of 8-bit input images it will result in truncated derivatives.

  • dx: int.

    order of the derivative x.

  • dy: int.

    order of the derivative y.

Keyword Arguments
  • ksize: int.

    size of the extended Sobel kernel; it must be 1, 3, 5, or 7.

  • scale: double.

    optional scale factor for the computed derivative values; by default, no scaling is applied (see #getDerivKernels for details).

  • delta: double.

    optional delta value that is added to the results prior to storing them in dst.

  • borderType: int.

    pixel extrapolation method, see #BorderTypes. #BORDER_WRAP is not supported.

Return
  • dst: Evision.Mat.t().

    output image of the same size and the same number of channels as src .

In all cases except one, the \f$\texttt{ksize} \times \texttt{ksize}\f$ separable kernel is used to calculate the derivative. When \f$\texttt{ksize = 1}\f$, the \f$3 \times 1\f$ or \f$1 \times 3\f$ kernel is used (that is, no Gaussian smoothing is done). ksize = 1 can only be used for the first or the second x- or y- derivatives. There is also the special value ksize = #FILTER_SCHARR (-1) that corresponds to the \f$3\times3\f$ Scharr filter that may give more accurate results than the \f$3\times3\f$ Sobel. The Scharr aperture is \f[\vecthreethree{-3}{0}{3}{-10}{0}{10}{-3}{0}{3}\f] for the x-derivative, or transposed for the y-derivative. The function calculates an image derivative by convolving the image with the appropriate kernel: \f[\texttt{dst} = \frac{\partial^{xorder+yorder} \texttt{src}}{\partial x^{xorder} \partial y^{yorder}}\f] The Sobel operators combine Gaussian smoothing and differentiation, so the result is more or less resistant to the noise. Most often, the function is called with ( xorder = 1, yorder = 0, ksize = 3) or ( xorder = 0, yorder = 1, ksize = 3) to calculate the first x- or y- image derivative. The first case corresponds to a kernel of: \f[\vecthreethree{-1}{0}{1}{-2}{0}{2}{-1}{0}{1}\f] The second case corresponds to a kernel of: \f[\vecthreethree{-1}{-2}{-1}{0}{0}{0}{1}{2}{1}\f] @sa Scharr, Laplacian, sepFilter2D, filter2D, GaussianBlur, cartToPolar

Python prototype (for reference only):

Sobel(src, ddepth, dx, dy[, dst[, ksize[, scale[, delta[, borderType]]]]]) -> dst
@spec solve(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | false | {:error, String.t()}

Solves one or more linear systems or least-squares problems.

Positional Arguments
  • src1: Evision.Mat.t().

    input matrix on the left-hand side of the system.

  • src2: Evision.Mat.t().

    input matrix on the right-hand side of the system.

Keyword Arguments
  • flags: int.

    solution (matrix inversion) method (#DecompTypes)

Return
  • retval: bool

  • dst: Evision.Mat.t().

    output solution.

The function cv::solve solves a linear system or least-squares problem (the latter is possible with SVD or QR methods, or by specifying the flag #DECOMP_NORMAL ): \f[\texttt{dst} = \arg \min _X \| \texttt{src1} \cdot \texttt{X} - \texttt{src2} \|\f] If #DECOMP_LU or #DECOMP_CHOLESKY method is used, the function returns 1 if src1 (or \f$\texttt{src1}^T\texttt{src1}\f$ ) is non-singular. Otherwise, it returns 0. In the latter case, dst is not valid. Other methods find a pseudo-solution in case of a singular left-hand side part. Note: If you want to find a unity-norm solution of an under-defined singular system \f$\texttt{src1}\cdot\texttt{dst}=0\f$ , the function solve will not do the work. Use SVD::solveZ instead. @sa invert, SVD, eigen

Python prototype (for reference only):

solve(src1, src2[, dst[, flags]]) -> retval, dst
@spec solve(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | false | {:error, String.t()}

Solves one or more linear systems or least-squares problems.

Positional Arguments
  • src1: Evision.Mat.t().

    input matrix on the left-hand side of the system.

  • src2: Evision.Mat.t().

    input matrix on the right-hand side of the system.

Keyword Arguments
  • flags: int.

    solution (matrix inversion) method (#DecompTypes)

Return
  • retval: bool

  • dst: Evision.Mat.t().

    output solution.

The function cv::solve solves a linear system or least-squares problem (the latter is possible with SVD or QR methods, or by specifying the flag #DECOMP_NORMAL ): \f[\texttt{dst} = \arg \min _X \| \texttt{src1} \cdot \texttt{X} - \texttt{src2} \|\f] If #DECOMP_LU or #DECOMP_CHOLESKY method is used, the function returns 1 if src1 (or \f$\texttt{src1}^T\texttt{src1}\f$ ) is non-singular. Otherwise, it returns 0. In the latter case, dst is not valid. Other methods find a pseudo-solution in case of a singular left-hand side part. Note: If you want to find a unity-norm solution of an under-defined singular system \f$\texttt{src1}\cdot\texttt{dst}=0\f$ , the function solve will not do the work. Use SVD::solveZ instead. @sa invert, SVD, eigen

Python prototype (for reference only):

solve(src1, src2[, dst[, flags]]) -> retval, dst
@spec solveCubic(Evision.Mat.maybe_mat_in()) ::
  {integer(), Evision.Mat.t()} | {:error, String.t()}

Finds the real roots of a cubic equation.

Positional Arguments
  • coeffs: Evision.Mat.t().

    equation coefficients, an array of 3 or 4 elements.

Return
  • retval: int

  • roots: Evision.Mat.t().

    output array of real roots that has 1 or 3 elements.

The function solveCubic finds the real roots of a cubic equation:

  • if coeffs is a 4-element vector: \f[\texttt{coeffs} [0] x^3 + \texttt{coeffs} [1] x^2 + \texttt{coeffs} [2] x + \texttt{coeffs} [3] = 0\f]

  • if coeffs is a 3-element vector: \f[x^3 + \texttt{coeffs} [0] x^2 + \texttt{coeffs} [1] x + \texttt{coeffs} [2] = 0\f]

The roots are stored in the roots array. @return number of real roots. It can be 0, 1 or 2.

Python prototype (for reference only):

solveCubic(coeffs[, roots]) -> retval, roots
Link to this function

solveCubic(coeffs, opts)

View Source
@spec solveCubic(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  {integer(), Evision.Mat.t()} | {:error, String.t()}

Finds the real roots of a cubic equation.

Positional Arguments
  • coeffs: Evision.Mat.t().

    equation coefficients, an array of 3 or 4 elements.

Return
  • retval: int

  • roots: Evision.Mat.t().

    output array of real roots that has 1 or 3 elements.

The function solveCubic finds the real roots of a cubic equation:

  • if coeffs is a 4-element vector: \f[\texttt{coeffs} [0] x^3 + \texttt{coeffs} [1] x^2 + \texttt{coeffs} [2] x + \texttt{coeffs} [3] = 0\f]

  • if coeffs is a 3-element vector: \f[x^3 + \texttt{coeffs} [0] x^2 + \texttt{coeffs} [1] x + \texttt{coeffs} [2] = 0\f]

The roots are stored in the roots array. @return number of real roots. It can be 0, 1 or 2.

Python prototype (for reference only):

solveCubic(coeffs[, roots]) -> retval, roots

solveLP

Positional Arguments
  • func: Evision.Mat.t()
  • constr: Evision.Mat.t()
Return
  • retval: int
  • z: Evision.Mat.t().

Has overloading in C++

Python prototype (for reference only):

solveLP(Func, Constr[, z]) -> retval, z
Link to this function

solveLP(func, constr, opts)

View Source
@spec solveLP(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  {integer(), Evision.Mat.t()} | {:error, String.t()}
@spec solveLP(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in(), number()) ::
  {integer(), Evision.Mat.t()} | {:error, String.t()}

Variant 1:

Solve given (non-integer) linear programming problem using the Simplex Algorithm (Simplex Method).

Positional Arguments
  • func: Evision.Mat.t().

    This row-vector corresponds to \f$c\f$ in the LP problem formulation (see above). It should contain 32- or 64-bit floating point numbers. As a convenience, column-vector may be also submitted, in the latter case it is understood to correspond to \f$c^T\f$.

  • constr: Evision.Mat.t().

    m-by-n+1 matrix, whose rightmost column corresponds to \f$b\f$ in formulation above and the remaining to \f$A\f$. It should contain 32- or 64-bit floating point numbers.

  • constr_eps: double.

    allowed numeric disparity for constraints

Return
  • retval: int

  • z: Evision.Mat.t().

    The solution will be returned here as a column-vector - it corresponds to \f$c\f$ in the formulation above. It will contain 64-bit floating point numbers.

What we mean here by "linear programming problem" (or LP problem, for short) can be formulated as: \f[\mbox{Maximize } c\cdot x\\ \mbox{Subject to:}\\ Ax\leq b\\ x\geq 0\f] Where \f$c\f$ is fixed 1-by-n row-vector, \f$A\f$ is fixed m-by-n matrix, \f$b\f$ is fixed m-by-1 column vector and \f$x\f$ is an arbitrary n-by-1 column vector, which satisfies the constraints. Simplex algorithm is one of many algorithms that are designed to handle this sort of problems efficiently. Although it is not optimal in theoretical sense (there exist algorithms that can solve any problem written as above in polynomial time, while simplex method degenerates to exponential time for some special cases), it is well-studied, easy to implement and is shown to work well for real-life purposes. The particular implementation is taken almost verbatim from Introduction to Algorithms, third edition by T. H. Cormen, C. E. Leiserson, R. L. Rivest and Clifford Stein. In particular, the Bland's rule http://en.wikipedia.org/wiki/Bland%27s_rule is used to prevent cycling. @return One of cv::SolveLPResult

Python prototype (for reference only):

solveLP(Func, Constr, constr_eps[, z]) -> retval, z

Variant 2:

solveLP

Positional Arguments
  • func: Evision.Mat.t()
  • constr: Evision.Mat.t()
Return
  • retval: int
  • z: Evision.Mat.t().

Has overloading in C++

Python prototype (for reference only):

solveLP(Func, Constr[, z]) -> retval, z
Link to this function

solveLP(func, constr, constr_eps, opts)

View Source
@spec solveLP(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  number(),
  [{atom(), term()}, ...] | nil
) :: {integer(), Evision.Mat.t()} | {:error, String.t()}

Solve given (non-integer) linear programming problem using the Simplex Algorithm (Simplex Method).

Positional Arguments
  • func: Evision.Mat.t().

    This row-vector corresponds to \f$c\f$ in the LP problem formulation (see above). It should contain 32- or 64-bit floating point numbers. As a convenience, column-vector may be also submitted, in the latter case it is understood to correspond to \f$c^T\f$.

  • constr: Evision.Mat.t().

    m-by-n+1 matrix, whose rightmost column corresponds to \f$b\f$ in formulation above and the remaining to \f$A\f$. It should contain 32- or 64-bit floating point numbers.

  • constr_eps: double.

    allowed numeric disparity for constraints

Return
  • retval: int

  • z: Evision.Mat.t().

    The solution will be returned here as a column-vector - it corresponds to \f$c\f$ in the formulation above. It will contain 64-bit floating point numbers.

What we mean here by "linear programming problem" (or LP problem, for short) can be formulated as: \f[\mbox{Maximize } c\cdot x\\ \mbox{Subject to:}\\ Ax\leq b\\ x\geq 0\f] Where \f$c\f$ is fixed 1-by-n row-vector, \f$A\f$ is fixed m-by-n matrix, \f$b\f$ is fixed m-by-1 column vector and \f$x\f$ is an arbitrary n-by-1 column vector, which satisfies the constraints. Simplex algorithm is one of many algorithms that are designed to handle this sort of problems efficiently. Although it is not optimal in theoretical sense (there exist algorithms that can solve any problem written as above in polynomial time, while simplex method degenerates to exponential time for some special cases), it is well-studied, easy to implement and is shown to work well for real-life purposes. The particular implementation is taken almost verbatim from Introduction to Algorithms, third edition by T. H. Cormen, C. E. Leiserson, R. L. Rivest and Clifford Stein. In particular, the Bland's rule http://en.wikipedia.org/wiki/Bland%27s_rule is used to prevent cycling. @return One of cv::SolveLPResult

Python prototype (for reference only):

solveLP(Func, Constr, constr_eps[, z]) -> retval, z
Link to this function

solveP3P(objectPoints, imagePoints, cameraMatrix, distCoeffs, flags)

View Source

Finds an object pose from 3 3D-2D point correspondences.

Positional Arguments
  • objectPoints: Evision.Mat.t().

    Array of object points in the object coordinate space, 3x3 1-channel or 1x3/3x1 3-channel. vector\<Point3f> can be also passed here.

  • imagePoints: Evision.Mat.t().

    Array of corresponding image points, 3x2 1-channel or 1x3/3x1 2-channel. vector\<Point2f> can be also passed here.

  • cameraMatrix: Evision.Mat.t().

    Input camera intrinsic matrix \f$\cameramatrix{A}\f$ .

  • distCoeffs: Evision.Mat.t().

    Input vector of distortion coefficients \f$\distcoeffs\f$. If the vector is NULL/empty, the zero distortion coefficients are assumed.

  • flags: int.

    Method for solving a P3P problem:

    • @ref SOLVEPNP_P3P Method is based on the paper of X.S. Gao, X.-R. Hou, J. Tang, H.-F. Chang "Complete Solution Classification for the Perspective-Three-Point Problem" (@cite gao2003complete).
    • @ref SOLVEPNP_AP3P Method is based on the paper of T. Ke and S. Roumeliotis. "An Efficient Algebraic Solution to the Perspective-Three-Point Problem" (@cite Ke17).
Return
  • retval: int

  • rvecs: [Evision.Mat].

    Output rotation vectors (see @ref Rodrigues ) that, together with tvecs, brings points from the model coordinate system to the camera coordinate system. A P3P problem has up to 4 solutions.

  • tvecs: [Evision.Mat].

    Output translation vectors.

@see @ref calib3d_solvePnP

The function estimates the object pose given 3 object points, their corresponding image projections, as well as the camera intrinsic matrix and the distortion coefficients. Note: The solutions are sorted by reprojection errors (lowest to highest).

Python prototype (for reference only):

solveP3P(objectPoints, imagePoints, cameraMatrix, distCoeffs, flags[, rvecs[, tvecs]]) -> retval, rvecs, tvecs
Link to this function

solveP3P(objectPoints, imagePoints, cameraMatrix, distCoeffs, flags, opts)

View Source

Finds an object pose from 3 3D-2D point correspondences.

Positional Arguments
  • objectPoints: Evision.Mat.t().

    Array of object points in the object coordinate space, 3x3 1-channel or 1x3/3x1 3-channel. vector\<Point3f> can be also passed here.

  • imagePoints: Evision.Mat.t().

    Array of corresponding image points, 3x2 1-channel or 1x3/3x1 2-channel. vector\<Point2f> can be also passed here.

  • cameraMatrix: Evision.Mat.t().

    Input camera intrinsic matrix \f$\cameramatrix{A}\f$ .

  • distCoeffs: Evision.Mat.t().

    Input vector of distortion coefficients \f$\distcoeffs\f$. If the vector is NULL/empty, the zero distortion coefficients are assumed.

  • flags: int.

    Method for solving a P3P problem:

    • @ref SOLVEPNP_P3P Method is based on the paper of X.S. Gao, X.-R. Hou, J. Tang, H.-F. Chang "Complete Solution Classification for the Perspective-Three-Point Problem" (@cite gao2003complete).
    • @ref SOLVEPNP_AP3P Method is based on the paper of T. Ke and S. Roumeliotis. "An Efficient Algebraic Solution to the Perspective-Three-Point Problem" (@cite Ke17).
Return
  • retval: int

  • rvecs: [Evision.Mat].

    Output rotation vectors (see @ref Rodrigues ) that, together with tvecs, brings points from the model coordinate system to the camera coordinate system. A P3P problem has up to 4 solutions.

  • tvecs: [Evision.Mat].

    Output translation vectors.

@see @ref calib3d_solvePnP

The function estimates the object pose given 3 object points, their corresponding image projections, as well as the camera intrinsic matrix and the distortion coefficients. Note: The solutions are sorted by reprojection errors (lowest to highest).

Python prototype (for reference only):

solveP3P(objectPoints, imagePoints, cameraMatrix, distCoeffs, flags[, rvecs[, tvecs]]) -> retval, rvecs, tvecs
Link to this function

solvePnP(objectPoints, imagePoints, cameraMatrix, distCoeffs)

View Source

Finds an object pose from 3D-2D point correspondences.

Positional Arguments
  • objectPoints: Evision.Mat.t().

    Array of object points in the object coordinate space, Nx3 1-channel or 1xN/Nx1 3-channel, where N is the number of points. vector\<Point3d> can be also passed here.

  • imagePoints: Evision.Mat.t().

    Array of corresponding image points, Nx2 1-channel or 1xN/Nx1 2-channel, where N is the number of points. vector\<Point2d> can be also passed here.

  • cameraMatrix: Evision.Mat.t().

    Input camera intrinsic matrix \f$\cameramatrix{A}\f$ .

  • distCoeffs: Evision.Mat.t().

    Input vector of distortion coefficients \f$\distcoeffs\f$. If the vector is NULL/empty, the zero distortion coefficients are assumed.

Keyword Arguments
  • useExtrinsicGuess: bool.

    Parameter used for #SOLVEPNP_ITERATIVE. If true (1), the function uses the provided rvec and tvec values as initial approximations of the rotation and translation vectors, respectively, and further optimizes them.

  • flags: int.

    Method for solving a PnP problem: see @ref calib3d_solvePnP_flags

Return
  • retval: bool

  • rvec: Evision.Mat.t().

    Output rotation vector (see @ref Rodrigues ) that, together with tvec, brings points from the model coordinate system to the camera coordinate system.

  • tvec: Evision.Mat.t().

    Output translation vector.

@see @ref calib3d_solvePnP This function returns the rotation and the translation vectors that transform a 3D point expressed in the object coordinate frame to the camera coordinate frame, using different methods:

  • P3P methods (@ref SOLVEPNP_P3P, @ref SOLVEPNP_AP3P): need 4 input points to return a unique solution.

  • @ref SOLVEPNP_IPPE Input points must be >= 4 and object points must be coplanar.

  • @ref SOLVEPNP_IPPE_SQUARE Special case suitable for marker pose estimation. Number of input points must be 4. Object points must be defined in the following order:

  • point 0: [-squareLength / 2, squareLength / 2, 0]

  • point 1: [ squareLength / 2, squareLength / 2, 0]

  • point 2: [ squareLength / 2, -squareLength / 2, 0]

  • point 3: [-squareLength / 2, -squareLength / 2, 0]

  • for all the other flags, number of input points must be >= 4 and object points can be in any configuration.

More information about Perspective-n-Points is described in @ref calib3d_solvePnP Note:

  • An example of how to use solvePnP for planar augmented reality can be found at opencv_source_code/samples/python/plane_ar.py

  • If you are using Python:

  • Numpy array slices won't work as input because solvePnP requires contiguous arrays (enforced by the assertion using cv::Mat::checkVector() around line 55 of modules/calib3d/src/solvepnp.cpp version 2.4.9)

  • The P3P algorithm requires image points to be in an array of shape (N,1,2) due to its calling of #undistortPoints (around line 75 of modules/calib3d/src/solvepnp.cpp version 2.4.9) which requires 2-channel information.

  • Thus, given some data D = np.array(...) where D.shape = (N,M), in order to use a subset of it as, e.g., imagePoints, one must effectively copy it into a new array: imagePoints = np.ascontiguousarray(D[:,:2]).reshape((N,1,2))

  • The methods @ref SOLVEPNP_DLS and @ref SOLVEPNP_UPNP cannot be used as the current implementations are unstable and sometimes give completely wrong results. If you pass one of these two flags, @ref SOLVEPNP_EPNP method will be used instead.

  • The minimum number of points is 4 in the general case. In the case of @ref SOLVEPNP_P3P and @ref SOLVEPNP_AP3P methods, it is required to use exactly 4 points (the first 3 points are used to estimate all the solutions of the P3P problem, the last one is used to retain the best solution that minimizes the reprojection error).

  • With @ref SOLVEPNP_ITERATIVE method and useExtrinsicGuess=true, the minimum number of points is 3 (3 points are sufficient to compute a pose but there are up to 4 solutions). The initial solution should be close to the global solution to converge.

  • With @ref SOLVEPNP_IPPE input points must be >= 4 and object points must be coplanar.

  • With @ref SOLVEPNP_IPPE_SQUARE this is a special case suitable for marker pose estimation. Number of input points must be 4. Object points must be defined in the following order:

  • point 0: [-squareLength / 2, squareLength / 2, 0]

  • point 1: [ squareLength / 2, squareLength / 2, 0]

  • point 2: [ squareLength / 2, -squareLength / 2, 0]

  • point 3: [-squareLength / 2, -squareLength / 2, 0]

  • With @ref SOLVEPNP_SQPNP input points must be >= 3

Python prototype (for reference only):

solvePnP(objectPoints, imagePoints, cameraMatrix, distCoeffs[, rvec[, tvec[, useExtrinsicGuess[, flags]]]]) -> retval, rvec, tvec
Link to this function

solvePnP(objectPoints, imagePoints, cameraMatrix, distCoeffs, opts)

View Source
@spec solvePnP(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: {Evision.Mat.t(), Evision.Mat.t()} | false | {:error, String.t()}

Finds an object pose from 3D-2D point correspondences.

Positional Arguments
  • objectPoints: Evision.Mat.t().

    Array of object points in the object coordinate space, Nx3 1-channel or 1xN/Nx1 3-channel, where N is the number of points. vector\<Point3d> can be also passed here.

  • imagePoints: Evision.Mat.t().

    Array of corresponding image points, Nx2 1-channel or 1xN/Nx1 2-channel, where N is the number of points. vector\<Point2d> can be also passed here.

  • cameraMatrix: Evision.Mat.t().

    Input camera intrinsic matrix \f$\cameramatrix{A}\f$ .

  • distCoeffs: Evision.Mat.t().

    Input vector of distortion coefficients \f$\distcoeffs\f$. If the vector is NULL/empty, the zero distortion coefficients are assumed.

Keyword Arguments
  • useExtrinsicGuess: bool.

    Parameter used for #SOLVEPNP_ITERATIVE. If true (1), the function uses the provided rvec and tvec values as initial approximations of the rotation and translation vectors, respectively, and further optimizes them.

  • flags: int.

    Method for solving a PnP problem: see @ref calib3d_solvePnP_flags

Return
  • retval: bool

  • rvec: Evision.Mat.t().

    Output rotation vector (see @ref Rodrigues ) that, together with tvec, brings points from the model coordinate system to the camera coordinate system.

  • tvec: Evision.Mat.t().

    Output translation vector.

@see @ref calib3d_solvePnP This function returns the rotation and the translation vectors that transform a 3D point expressed in the object coordinate frame to the camera coordinate frame, using different methods:

  • P3P methods (@ref SOLVEPNP_P3P, @ref SOLVEPNP_AP3P): need 4 input points to return a unique solution.

  • @ref SOLVEPNP_IPPE Input points must be >= 4 and object points must be coplanar.

  • @ref SOLVEPNP_IPPE_SQUARE Special case suitable for marker pose estimation. Number of input points must be 4. Object points must be defined in the following order:

  • point 0: [-squareLength / 2, squareLength / 2, 0]

  • point 1: [ squareLength / 2, squareLength / 2, 0]

  • point 2: [ squareLength / 2, -squareLength / 2, 0]

  • point 3: [-squareLength / 2, -squareLength / 2, 0]

  • for all the other flags, number of input points must be >= 4 and object points can be in any configuration.

More information about Perspective-n-Points is described in @ref calib3d_solvePnP Note:

  • An example of how to use solvePnP for planar augmented reality can be found at opencv_source_code/samples/python/plane_ar.py

  • If you are using Python:

  • Numpy array slices won't work as input because solvePnP requires contiguous arrays (enforced by the assertion using cv::Mat::checkVector() around line 55 of modules/calib3d/src/solvepnp.cpp version 2.4.9)

  • The P3P algorithm requires image points to be in an array of shape (N,1,2) due to its calling of #undistortPoints (around line 75 of modules/calib3d/src/solvepnp.cpp version 2.4.9) which requires 2-channel information.

  • Thus, given some data D = np.array(...) where D.shape = (N,M), in order to use a subset of it as, e.g., imagePoints, one must effectively copy it into a new array: imagePoints = np.ascontiguousarray(D[:,:2]).reshape((N,1,2))

  • The methods @ref SOLVEPNP_DLS and @ref SOLVEPNP_UPNP cannot be used as the current implementations are unstable and sometimes give completely wrong results. If you pass one of these two flags, @ref SOLVEPNP_EPNP method will be used instead.

  • The minimum number of points is 4 in the general case. In the case of @ref SOLVEPNP_P3P and @ref SOLVEPNP_AP3P methods, it is required to use exactly 4 points (the first 3 points are used to estimate all the solutions of the P3P problem, the last one is used to retain the best solution that minimizes the reprojection error).

  • With @ref SOLVEPNP_ITERATIVE method and useExtrinsicGuess=true, the minimum number of points is 3 (3 points are sufficient to compute a pose but there are up to 4 solutions). The initial solution should be close to the global solution to converge.

  • With @ref SOLVEPNP_IPPE input points must be >= 4 and object points must be coplanar.

  • With @ref SOLVEPNP_IPPE_SQUARE this is a special case suitable for marker pose estimation. Number of input points must be 4. Object points must be defined in the following order:

  • point 0: [-squareLength / 2, squareLength / 2, 0]

  • point 1: [ squareLength / 2, squareLength / 2, 0]

  • point 2: [ squareLength / 2, -squareLength / 2, 0]

  • point 3: [-squareLength / 2, -squareLength / 2, 0]

  • With @ref SOLVEPNP_SQPNP input points must be >= 3

Python prototype (for reference only):

solvePnP(objectPoints, imagePoints, cameraMatrix, distCoeffs[, rvec[, tvec[, useExtrinsicGuess[, flags]]]]) -> retval, rvec, tvec
Link to this function

solvePnPGeneric(objectPoints, imagePoints, cameraMatrix, distCoeffs)

View Source

Finds an object pose from 3D-2D point correspondences.

Positional Arguments
  • objectPoints: Evision.Mat.t().

    Array of object points in the object coordinate space, Nx3 1-channel or 1xN/Nx1 3-channel, where N is the number of points. vector\<Point3d> can be also passed here.

  • imagePoints: Evision.Mat.t().

    Array of corresponding image points, Nx2 1-channel or 1xN/Nx1 2-channel, where N is the number of points. vector\<Point2d> can be also passed here.

  • cameraMatrix: Evision.Mat.t().

    Input camera intrinsic matrix \f$\cameramatrix{A}\f$ .

  • distCoeffs: Evision.Mat.t().

    Input vector of distortion coefficients \f$\distcoeffs\f$. If the vector is NULL/empty, the zero distortion coefficients are assumed.

Keyword Arguments
  • useExtrinsicGuess: bool.

    Parameter used for #SOLVEPNP_ITERATIVE. If true (1), the function uses the provided rvec and tvec values as initial approximations of the rotation and translation vectors, respectively, and further optimizes them.

  • flags: SolvePnPMethod.

    Method for solving a PnP problem: see @ref calib3d_solvePnP_flags

  • rvec: Evision.Mat.t().

    Rotation vector used to initialize an iterative PnP refinement algorithm, when flag is @ref SOLVEPNP_ITERATIVE and useExtrinsicGuess is set to true.

  • tvec: Evision.Mat.t().

    Translation vector used to initialize an iterative PnP refinement algorithm, when flag is @ref SOLVEPNP_ITERATIVE and useExtrinsicGuess is set to true.

Return
  • retval: int

  • rvecs: [Evision.Mat].

    Vector of output rotation vectors (see @ref Rodrigues ) that, together with tvecs, brings points from the model coordinate system to the camera coordinate system.

  • tvecs: [Evision.Mat].

    Vector of output translation vectors.

  • reprojectionError: Evision.Mat.t().

    Optional vector of reprojection error, that is the RMS error (\f$ \text{RMSE} = \sqrt{\frac{\sum_{i}^{N} \left ( \hat{y_i} - y_i \right )^2}{N}} \f$) between the input image points and the 3D object points projected with the estimated pose.

@see @ref calib3d_solvePnP This function returns a list of all the possible solutions (a solution is a <rotation vector, translation vector> couple), depending on the number of input points and the chosen method:

  • P3P methods (@ref SOLVEPNP_P3P, @ref SOLVEPNP_AP3P): 3 or 4 input points. Number of returned solutions can be between 0 and 4 with 3 input points.

  • @ref SOLVEPNP_IPPE Input points must be >= 4 and object points must be coplanar. Returns 2 solutions.

  • @ref SOLVEPNP_IPPE_SQUARE Special case suitable for marker pose estimation. Number of input points must be 4 and 2 solutions are returned. Object points must be defined in the following order:

  • point 0: [-squareLength / 2, squareLength / 2, 0]

  • point 1: [ squareLength / 2, squareLength / 2, 0]

  • point 2: [ squareLength / 2, -squareLength / 2, 0]

  • point 3: [-squareLength / 2, -squareLength / 2, 0]

  • for all the other flags, number of input points must be >= 4 and object points can be in any configuration. Only 1 solution is returned.

More information is described in @ref calib3d_solvePnP Note:

  • An example of how to use solvePnP for planar augmented reality can be found at opencv_source_code/samples/python/plane_ar.py

  • If you are using Python:

  • Numpy array slices won't work as input because solvePnP requires contiguous arrays (enforced by the assertion using cv::Mat::checkVector() around line 55 of modules/calib3d/src/solvepnp.cpp version 2.4.9)

  • The P3P algorithm requires image points to be in an array of shape (N,1,2) due to its calling of #undistortPoints (around line 75 of modules/calib3d/src/solvepnp.cpp version 2.4.9) which requires 2-channel information.

  • Thus, given some data D = np.array(...) where D.shape = (N,M), in order to use a subset of it as, e.g., imagePoints, one must effectively copy it into a new array: imagePoints = np.ascontiguousarray(D[:,:2]).reshape((N,1,2))

  • The methods @ref SOLVEPNP_DLS and @ref SOLVEPNP_UPNP cannot be used as the current implementations are unstable and sometimes give completely wrong results. If you pass one of these two flags, @ref SOLVEPNP_EPNP method will be used instead.

  • The minimum number of points is 4 in the general case. In the case of @ref SOLVEPNP_P3P and @ref SOLVEPNP_AP3P methods, it is required to use exactly 4 points (the first 3 points are used to estimate all the solutions of the P3P problem, the last one is used to retain the best solution that minimizes the reprojection error).

  • With @ref SOLVEPNP_ITERATIVE method and useExtrinsicGuess=true, the minimum number of points is 3 (3 points are sufficient to compute a pose but there are up to 4 solutions). The initial solution should be close to the global solution to converge.

  • With @ref SOLVEPNP_IPPE input points must be >= 4 and object points must be coplanar.

  • With @ref SOLVEPNP_IPPE_SQUARE this is a special case suitable for marker pose estimation. Number of input points must be 4. Object points must be defined in the following order:

  • point 0: [-squareLength / 2, squareLength / 2, 0]

  • point 1: [ squareLength / 2, squareLength / 2, 0]

  • point 2: [ squareLength / 2, -squareLength / 2, 0]

  • point 3: [-squareLength / 2, -squareLength / 2, 0]

Python prototype (for reference only):

solvePnPGeneric(objectPoints, imagePoints, cameraMatrix, distCoeffs[, rvecs[, tvecs[, useExtrinsicGuess[, flags[, rvec[, tvec[, reprojectionError]]]]]]]) -> retval, rvecs, tvecs, reprojectionError
Link to this function

solvePnPGeneric(objectPoints, imagePoints, cameraMatrix, distCoeffs, opts)

View Source

Finds an object pose from 3D-2D point correspondences.

Positional Arguments
  • objectPoints: Evision.Mat.t().

    Array of object points in the object coordinate space, Nx3 1-channel or 1xN/Nx1 3-channel, where N is the number of points. vector\<Point3d> can be also passed here.

  • imagePoints: Evision.Mat.t().

    Array of corresponding image points, Nx2 1-channel or 1xN/Nx1 2-channel, where N is the number of points. vector\<Point2d> can be also passed here.

  • cameraMatrix: Evision.Mat.t().

    Input camera intrinsic matrix \f$\cameramatrix{A}\f$ .

  • distCoeffs: Evision.Mat.t().

    Input vector of distortion coefficients \f$\distcoeffs\f$. If the vector is NULL/empty, the zero distortion coefficients are assumed.

Keyword Arguments
  • useExtrinsicGuess: bool.

    Parameter used for #SOLVEPNP_ITERATIVE. If true (1), the function uses the provided rvec and tvec values as initial approximations of the rotation and translation vectors, respectively, and further optimizes them.

  • flags: SolvePnPMethod.

    Method for solving a PnP problem: see @ref calib3d_solvePnP_flags

  • rvec: Evision.Mat.t().

    Rotation vector used to initialize an iterative PnP refinement algorithm, when flag is @ref SOLVEPNP_ITERATIVE and useExtrinsicGuess is set to true.

  • tvec: Evision.Mat.t().

    Translation vector used to initialize an iterative PnP refinement algorithm, when flag is @ref SOLVEPNP_ITERATIVE and useExtrinsicGuess is set to true.

Return
  • retval: int

  • rvecs: [Evision.Mat].

    Vector of output rotation vectors (see @ref Rodrigues ) that, together with tvecs, brings points from the model coordinate system to the camera coordinate system.

  • tvecs: [Evision.Mat].

    Vector of output translation vectors.

  • reprojectionError: Evision.Mat.t().

    Optional vector of reprojection error, that is the RMS error (\f$ \text{RMSE} = \sqrt{\frac{\sum_{i}^{N} \left ( \hat{y_i} - y_i \right )^2}{N}} \f$) between the input image points and the 3D object points projected with the estimated pose.

@see @ref calib3d_solvePnP This function returns a list of all the possible solutions (a solution is a <rotation vector, translation vector> couple), depending on the number of input points and the chosen method:

  • P3P methods (@ref SOLVEPNP_P3P, @ref SOLVEPNP_AP3P): 3 or 4 input points. Number of returned solutions can be between 0 and 4 with 3 input points.

  • @ref SOLVEPNP_IPPE Input points must be >= 4 and object points must be coplanar. Returns 2 solutions.

  • @ref SOLVEPNP_IPPE_SQUARE Special case suitable for marker pose estimation. Number of input points must be 4 and 2 solutions are returned. Object points must be defined in the following order:

  • point 0: [-squareLength / 2, squareLength / 2, 0]

  • point 1: [ squareLength / 2, squareLength / 2, 0]

  • point 2: [ squareLength / 2, -squareLength / 2, 0]

  • point 3: [-squareLength / 2, -squareLength / 2, 0]

  • for all the other flags, number of input points must be >= 4 and object points can be in any configuration. Only 1 solution is returned.

More information is described in @ref calib3d_solvePnP Note:

  • An example of how to use solvePnP for planar augmented reality can be found at opencv_source_code/samples/python/plane_ar.py

  • If you are using Python:

  • Numpy array slices won't work as input because solvePnP requires contiguous arrays (enforced by the assertion using cv::Mat::checkVector() around line 55 of modules/calib3d/src/solvepnp.cpp version 2.4.9)

  • The P3P algorithm requires image points to be in an array of shape (N,1,2) due to its calling of #undistortPoints (around line 75 of modules/calib3d/src/solvepnp.cpp version 2.4.9) which requires 2-channel information.

  • Thus, given some data D = np.array(...) where D.shape = (N,M), in order to use a subset of it as, e.g., imagePoints, one must effectively copy it into a new array: imagePoints = np.ascontiguousarray(D[:,:2]).reshape((N,1,2))

  • The methods @ref SOLVEPNP_DLS and @ref SOLVEPNP_UPNP cannot be used as the current implementations are unstable and sometimes give completely wrong results. If you pass one of these two flags, @ref SOLVEPNP_EPNP method will be used instead.

  • The minimum number of points is 4 in the general case. In the case of @ref SOLVEPNP_P3P and @ref SOLVEPNP_AP3P methods, it is required to use exactly 4 points (the first 3 points are used to estimate all the solutions of the P3P problem, the last one is used to retain the best solution that minimizes the reprojection error).

  • With @ref SOLVEPNP_ITERATIVE method and useExtrinsicGuess=true, the minimum number of points is 3 (3 points are sufficient to compute a pose but there are up to 4 solutions). The initial solution should be close to the global solution to converge.

  • With @ref SOLVEPNP_IPPE input points must be >= 4 and object points must be coplanar.

  • With @ref SOLVEPNP_IPPE_SQUARE this is a special case suitable for marker pose estimation. Number of input points must be 4. Object points must be defined in the following order:

  • point 0: [-squareLength / 2, squareLength / 2, 0]

  • point 1: [ squareLength / 2, squareLength / 2, 0]

  • point 2: [ squareLength / 2, -squareLength / 2, 0]

  • point 3: [-squareLength / 2, -squareLength / 2, 0]

Python prototype (for reference only):

solvePnPGeneric(objectPoints, imagePoints, cameraMatrix, distCoeffs[, rvecs[, tvecs[, useExtrinsicGuess[, flags[, rvec[, tvec[, reprojectionError]]]]]]]) -> retval, rvecs, tvecs, reprojectionError
Link to this function

solvePnPRansac(objectPoints, imagePoints, cameraMatrix, distCoeffs)

View Source

solvePnPRansac

Positional Arguments
  • objectPoints: Evision.Mat.t()
  • imagePoints: Evision.Mat.t()
  • distCoeffs: Evision.Mat.t()
Keyword Arguments
  • params: Evision.UsacParams.t().
Return
  • retval: bool
  • cameraMatrix: Evision.Mat.t()
  • rvec: Evision.Mat.t().
  • tvec: Evision.Mat.t().
  • inliers: Evision.Mat.t().

Python prototype (for reference only):

solvePnPRansac(objectPoints, imagePoints, cameraMatrix, distCoeffs[, rvec[, tvec[, inliers[, params]]]]) -> retval, cameraMatrix, rvec, tvec, inliers
Link to this function

solvePnPRansac(objectPoints, imagePoints, cameraMatrix, distCoeffs, opts)

View Source

solvePnPRansac

Positional Arguments
  • objectPoints: Evision.Mat.t()
  • imagePoints: Evision.Mat.t()
  • distCoeffs: Evision.Mat.t()
Keyword Arguments
  • params: Evision.UsacParams.t().
Return
  • retval: bool
  • cameraMatrix: Evision.Mat.t()
  • rvec: Evision.Mat.t().
  • tvec: Evision.Mat.t().
  • inliers: Evision.Mat.t().

Python prototype (for reference only):

solvePnPRansac(objectPoints, imagePoints, cameraMatrix, distCoeffs[, rvec[, tvec[, inliers[, params]]]]) -> retval, cameraMatrix, rvec, tvec, inliers
Link to this function

solvePnPRefineLM(objectPoints, imagePoints, cameraMatrix, distCoeffs, rvec, tvec)

View Source

Refine a pose (the translation and the rotation that transform a 3D point expressed in the object coordinate frame to the camera coordinate frame) from a 3D-2D point correspondences and starting from an initial solution.

Positional Arguments
  • objectPoints: Evision.Mat.t().

    Array of object points in the object coordinate space, Nx3 1-channel or 1xN/Nx1 3-channel, where N is the number of points. vector\<Point3d> can also be passed here.

  • imagePoints: Evision.Mat.t().

    Array of corresponding image points, Nx2 1-channel or 1xN/Nx1 2-channel, where N is the number of points. vector\<Point2d> can also be passed here.

  • cameraMatrix: Evision.Mat.t().

    Input camera intrinsic matrix \f$\cameramatrix{A}\f$ .

  • distCoeffs: Evision.Mat.t().

    Input vector of distortion coefficients \f$\distcoeffs\f$. If the vector is NULL/empty, the zero distortion coefficients are assumed.

Keyword Arguments
  • criteria: TermCriteria.

    Criteria when to stop the Levenberg-Marquard iterative algorithm.

Return
  • rvec: Evision.Mat.t().

    Input/Output rotation vector (see @ref Rodrigues ) that, together with tvec, brings points from the model coordinate system to the camera coordinate system. Input values are used as an initial solution.

  • tvec: Evision.Mat.t().

    Input/Output translation vector. Input values are used as an initial solution.

@see @ref calib3d_solvePnP

The function refines the object pose given at least 3 object points, their corresponding image projections, an initial solution for the rotation and translation vector, as well as the camera intrinsic matrix and the distortion coefficients. The function minimizes the projection error with respect to the rotation and the translation vectors, according to a Levenberg-Marquardt iterative minimization @cite Madsen04 @cite Eade13 process.

Python prototype (for reference only):

solvePnPRefineLM(objectPoints, imagePoints, cameraMatrix, distCoeffs, rvec, tvec[, criteria]) -> rvec, tvec
Link to this function

solvePnPRefineLM(objectPoints, imagePoints, cameraMatrix, distCoeffs, rvec, tvec, opts)

View Source

Refine a pose (the translation and the rotation that transform a 3D point expressed in the object coordinate frame to the camera coordinate frame) from a 3D-2D point correspondences and starting from an initial solution.

Positional Arguments
  • objectPoints: Evision.Mat.t().

    Array of object points in the object coordinate space, Nx3 1-channel or 1xN/Nx1 3-channel, where N is the number of points. vector\<Point3d> can also be passed here.

  • imagePoints: Evision.Mat.t().

    Array of corresponding image points, Nx2 1-channel or 1xN/Nx1 2-channel, where N is the number of points. vector\<Point2d> can also be passed here.

  • cameraMatrix: Evision.Mat.t().

    Input camera intrinsic matrix \f$\cameramatrix{A}\f$ .

  • distCoeffs: Evision.Mat.t().

    Input vector of distortion coefficients \f$\distcoeffs\f$. If the vector is NULL/empty, the zero distortion coefficients are assumed.

Keyword Arguments
  • criteria: TermCriteria.

    Criteria when to stop the Levenberg-Marquard iterative algorithm.

Return
  • rvec: Evision.Mat.t().

    Input/Output rotation vector (see @ref Rodrigues ) that, together with tvec, brings points from the model coordinate system to the camera coordinate system. Input values are used as an initial solution.

  • tvec: Evision.Mat.t().

    Input/Output translation vector. Input values are used as an initial solution.

@see @ref calib3d_solvePnP

The function refines the object pose given at least 3 object points, their corresponding image projections, an initial solution for the rotation and translation vector, as well as the camera intrinsic matrix and the distortion coefficients. The function minimizes the projection error with respect to the rotation and the translation vectors, according to a Levenberg-Marquardt iterative minimization @cite Madsen04 @cite Eade13 process.

Python prototype (for reference only):

solvePnPRefineLM(objectPoints, imagePoints, cameraMatrix, distCoeffs, rvec, tvec[, criteria]) -> rvec, tvec
Link to this function

solvePnPRefineVVS(objectPoints, imagePoints, cameraMatrix, distCoeffs, rvec, tvec)

View Source

Refine a pose (the translation and the rotation that transform a 3D point expressed in the object coordinate frame to the camera coordinate frame) from a 3D-2D point correspondences and starting from an initial solution.

Positional Arguments
  • objectPoints: Evision.Mat.t().

    Array of object points in the object coordinate space, Nx3 1-channel or 1xN/Nx1 3-channel, where N is the number of points. vector\<Point3d> can also be passed here.

  • imagePoints: Evision.Mat.t().

    Array of corresponding image points, Nx2 1-channel or 1xN/Nx1 2-channel, where N is the number of points. vector\<Point2d> can also be passed here.

  • cameraMatrix: Evision.Mat.t().

    Input camera intrinsic matrix \f$\cameramatrix{A}\f$ .

  • distCoeffs: Evision.Mat.t().

    Input vector of distortion coefficients \f$\distcoeffs\f$. If the vector is NULL/empty, the zero distortion coefficients are assumed.

Keyword Arguments
  • criteria: TermCriteria.

    Criteria when to stop the Levenberg-Marquard iterative algorithm.

  • vVSlambda: double.

    Gain for the virtual visual servoing control law, equivalent to the \f$\alpha\f$ gain in the Damped Gauss-Newton formulation.

Return
  • rvec: Evision.Mat.t().

    Input/Output rotation vector (see @ref Rodrigues ) that, together with tvec, brings points from the model coordinate system to the camera coordinate system. Input values are used as an initial solution.

  • tvec: Evision.Mat.t().

    Input/Output translation vector. Input values are used as an initial solution.

@see @ref calib3d_solvePnP

The function refines the object pose given at least 3 object points, their corresponding image projections, an initial solution for the rotation and translation vector, as well as the camera intrinsic matrix and the distortion coefficients. The function minimizes the projection error with respect to the rotation and the translation vectors, using a virtual visual servoing (VVS) @cite Chaumette06 @cite Marchand16 scheme.

Python prototype (for reference only):

solvePnPRefineVVS(objectPoints, imagePoints, cameraMatrix, distCoeffs, rvec, tvec[, criteria[, VVSlambda]]) -> rvec, tvec
Link to this function

solvePnPRefineVVS(objectPoints, imagePoints, cameraMatrix, distCoeffs, rvec, tvec, opts)

View Source

Refine a pose (the translation and the rotation that transform a 3D point expressed in the object coordinate frame to the camera coordinate frame) from a 3D-2D point correspondences and starting from an initial solution.

Positional Arguments
  • objectPoints: Evision.Mat.t().

    Array of object points in the object coordinate space, Nx3 1-channel or 1xN/Nx1 3-channel, where N is the number of points. vector\<Point3d> can also be passed here.

  • imagePoints: Evision.Mat.t().

    Array of corresponding image points, Nx2 1-channel or 1xN/Nx1 2-channel, where N is the number of points. vector\<Point2d> can also be passed here.

  • cameraMatrix: Evision.Mat.t().

    Input camera intrinsic matrix \f$\cameramatrix{A}\f$ .

  • distCoeffs: Evision.Mat.t().

    Input vector of distortion coefficients \f$\distcoeffs\f$. If the vector is NULL/empty, the zero distortion coefficients are assumed.

Keyword Arguments
  • criteria: TermCriteria.

    Criteria when to stop the Levenberg-Marquard iterative algorithm.

  • vVSlambda: double.

    Gain for the virtual visual servoing control law, equivalent to the \f$\alpha\f$ gain in the Damped Gauss-Newton formulation.

Return
  • rvec: Evision.Mat.t().

    Input/Output rotation vector (see @ref Rodrigues ) that, together with tvec, brings points from the model coordinate system to the camera coordinate system. Input values are used as an initial solution.

  • tvec: Evision.Mat.t().

    Input/Output translation vector. Input values are used as an initial solution.

@see @ref calib3d_solvePnP

The function refines the object pose given at least 3 object points, their corresponding image projections, an initial solution for the rotation and translation vector, as well as the camera intrinsic matrix and the distortion coefficients. The function minimizes the projection error with respect to the rotation and the translation vectors, using a virtual visual servoing (VVS) @cite Chaumette06 @cite Marchand16 scheme.

Python prototype (for reference only):

solvePnPRefineVVS(objectPoints, imagePoints, cameraMatrix, distCoeffs, rvec, tvec[, criteria[, VVSlambda]]) -> rvec, tvec
@spec solvePoly(Evision.Mat.maybe_mat_in()) ::
  {number(), Evision.Mat.t()} | {:error, String.t()}

Finds the real or complex roots of a polynomial equation.

Positional Arguments
  • coeffs: Evision.Mat.t().

    array of polynomial coefficients.

Keyword Arguments
  • maxIters: int.

    maximum number of iterations the algorithm does.

Return
  • retval: double

  • roots: Evision.Mat.t().

    output (complex) array of roots.

The function cv::solvePoly finds real and complex roots of a polynomial equation: \f[\texttt{coeffs} [n] x^{n} + \texttt{coeffs} [n-1] x^{n-1} + ... + \texttt{coeffs} [1] x + \texttt{coeffs} [0] = 0\f]

Python prototype (for reference only):

solvePoly(coeffs[, roots[, maxIters]]) -> retval, roots
@spec solvePoly(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  {number(), Evision.Mat.t()} | {:error, String.t()}

Finds the real or complex roots of a polynomial equation.

Positional Arguments
  • coeffs: Evision.Mat.t().

    array of polynomial coefficients.

Keyword Arguments
  • maxIters: int.

    maximum number of iterations the algorithm does.

Return
  • retval: double

  • roots: Evision.Mat.t().

    output (complex) array of roots.

The function cv::solvePoly finds real and complex roots of a polynomial equation: \f[\texttt{coeffs} [n] x^{n} + \texttt{coeffs} [n-1] x^{n-1} + ... + \texttt{coeffs} [1] x + \texttt{coeffs} [0] = 0\f]

Python prototype (for reference only):

solvePoly(coeffs[, roots[, maxIters]]) -> retval, roots
@spec sort(Evision.Mat.maybe_mat_in(), integer()) ::
  Evision.Mat.t() | {:error, String.t()}

Sorts each row or each column of a matrix.

Positional Arguments
  • src: Evision.Mat.t().

    input single-channel array.

  • flags: int.

    operation flags, a combination of #SortFlags

Return
  • dst: Evision.Mat.t().

    output array of the same size and type as src.

The function cv::sort sorts each matrix row or each matrix column in ascending or descending order. So you should pass two operation flags to get desired behaviour. If you want to sort matrix rows or columns lexicographically, you can use STL std::sort generic function with the proper comparison predicate. @sa sortIdx, randShuffle

Python prototype (for reference only):

sort(src, flags[, dst]) -> dst
@spec sort(Evision.Mat.maybe_mat_in(), integer(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Sorts each row or each column of a matrix.

Positional Arguments
  • src: Evision.Mat.t().

    input single-channel array.

  • flags: int.

    operation flags, a combination of #SortFlags

Return
  • dst: Evision.Mat.t().

    output array of the same size and type as src.

The function cv::sort sorts each matrix row or each matrix column in ascending or descending order. So you should pass two operation flags to get desired behaviour. If you want to sort matrix rows or columns lexicographically, you can use STL std::sort generic function with the proper comparison predicate. @sa sortIdx, randShuffle

Python prototype (for reference only):

sort(src, flags[, dst]) -> dst
@spec sortIdx(Evision.Mat.maybe_mat_in(), integer()) ::
  Evision.Mat.t() | {:error, String.t()}

Sorts each row or each column of a matrix.

Positional Arguments
  • src: Evision.Mat.t().

    input single-channel array.

  • flags: int.

    operation flags that could be a combination of cv::SortFlags

Return
  • dst: Evision.Mat.t().

    output integer array of the same size as src.

The function cv::sortIdx sorts each matrix row or each matrix column in the ascending or descending order. So you should pass two operation flags to get desired behaviour. Instead of reordering the elements themselves, it stores the indices of sorted elements in the output array. For example:

Mat A = Mat::eye(3,3,CV_32F), B;
sortIdx(A, B, SORT_EVERY_ROW + SORT_ASCENDING);
// B will probably contain
// (because of equal elements in A some permutations are possible):
// [[1, 2, 0], [0, 2, 1], [0, 1, 2]]

@sa sort, randShuffle

Python prototype (for reference only):

sortIdx(src, flags[, dst]) -> dst
Link to this function

sortIdx(src, flags, opts)

View Source
@spec sortIdx(Evision.Mat.maybe_mat_in(), integer(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Sorts each row or each column of a matrix.

Positional Arguments
  • src: Evision.Mat.t().

    input single-channel array.

  • flags: int.

    operation flags that could be a combination of cv::SortFlags

Return
  • dst: Evision.Mat.t().

    output integer array of the same size as src.

The function cv::sortIdx sorts each matrix row or each matrix column in the ascending or descending order. So you should pass two operation flags to get desired behaviour. Instead of reordering the elements themselves, it stores the indices of sorted elements in the output array. For example:

Mat A = Mat::eye(3,3,CV_32F), B;
sortIdx(A, B, SORT_EVERY_ROW + SORT_ASCENDING);
// B will probably contain
// (because of equal elements in A some permutations are possible):
// [[1, 2, 0], [0, 2, 1], [0, 1, 2]]

@sa sort, randShuffle

Python prototype (for reference only):

sortIdx(src, flags[, dst]) -> dst
@spec spatialGradient(Evision.Mat.maybe_mat_in()) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Calculates the first order image derivative in both x and y using a Sobel operator

Positional Arguments
  • src: Evision.Mat.t().

    input image.

Keyword Arguments
  • ksize: int.

    size of Sobel kernel. It must be 3.

  • borderType: int.

    pixel extrapolation method, see #BorderTypes. Only #BORDER_DEFAULT=#BORDER_REFLECT_101 and #BORDER_REPLICATE are supported.

Return
  • dx: Evision.Mat.t().

    output image with first-order derivative in x.

  • dy: Evision.Mat.t().

    output image with first-order derivative in y.

Equivalent to calling:

Sobel( src, dx, CV_16SC1, 1, 0, 3 );
Sobel( src, dy, CV_16SC1, 0, 1, 3 );

@sa Sobel

Python prototype (for reference only):

spatialGradient(src[, dx[, dy[, ksize[, borderType]]]]) -> dx, dy
Link to this function

spatialGradient(src, opts)

View Source
@spec spatialGradient(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  {Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

Calculates the first order image derivative in both x and y using a Sobel operator

Positional Arguments
  • src: Evision.Mat.t().

    input image.

Keyword Arguments
  • ksize: int.

    size of Sobel kernel. It must be 3.

  • borderType: int.

    pixel extrapolation method, see #BorderTypes. Only #BORDER_DEFAULT=#BORDER_REFLECT_101 and #BORDER_REPLICATE are supported.

Return
  • dx: Evision.Mat.t().

    output image with first-order derivative in x.

  • dy: Evision.Mat.t().

    output image with first-order derivative in y.

Equivalent to calling:

Sobel( src, dx, CV_16SC1, 1, 0, 3 );
Sobel( src, dy, CV_16SC1, 0, 1, 3 );

@sa Sobel

Python prototype (for reference only):

spatialGradient(src[, dx[, dy[, ksize[, borderType]]]]) -> dx, dy
@spec split(Evision.Mat.maybe_mat_in()) :: [Evision.Mat.t()] | {:error, String.t()}

split

Positional Arguments
  • m: Evision.Mat.t().

    input multi-channel array.

Return
  • mv: [Evision.Mat].

    output vector of arrays; the arrays themselves are reallocated, if needed.

Has overloading in C++

Python prototype (for reference only):

split(m[, mv]) -> mv
@spec split(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  [Evision.Mat.t()] | {:error, String.t()}

split

Positional Arguments
  • m: Evision.Mat.t().

    input multi-channel array.

Return
  • mv: [Evision.Mat].

    output vector of arrays; the arrays themselves are reallocated, if needed.

Has overloading in C++

Python prototype (for reference only):

split(m[, mv]) -> mv
Link to this function

sqrBoxFilter(src, ddepth, ksize)

View Source
@spec sqrBoxFilter(Evision.Mat.maybe_mat_in(), integer(), {number(), number()}) ::
  Evision.Mat.t() | {:error, String.t()}

Calculates the normalized sum of squares of the pixel values overlapping the filter.

Positional Arguments
  • src: Evision.Mat.t().

    input image

  • ddepth: int.

    the output image depth (-1 to use src.depth())

  • ksize: Size.

    kernel size

Keyword Arguments
  • anchor: Point.

    kernel anchor point. The default value of Point(-1, -1) denotes that the anchor is at the kernel center.

  • normalize: bool.

    flag, specifying whether the kernel is to be normalized by it's area or not.

  • borderType: int.

    border mode used to extrapolate pixels outside of the image, see #BorderTypes. #BORDER_WRAP is not supported.

Return
  • dst: Evision.Mat.t().

    output image of the same size and type as src

For every pixel \f$ (x, y) \f$ in the source image, the function calculates the sum of squares of those neighboring pixel values which overlap the filter placed over the pixel \f$ (x, y) \f$. The unnormalized square box filter can be useful in computing local image statistics such as the local variance and standard deviation around the neighborhood of a pixel. @sa boxFilter

Python prototype (for reference only):

sqrBoxFilter(src, ddepth, ksize[, dst[, anchor[, normalize[, borderType]]]]) -> dst
Link to this function

sqrBoxFilter(src, ddepth, ksize, opts)

View Source
@spec sqrBoxFilter(
  Evision.Mat.maybe_mat_in(),
  integer(),
  {number(), number()},
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Calculates the normalized sum of squares of the pixel values overlapping the filter.

Positional Arguments
  • src: Evision.Mat.t().

    input image

  • ddepth: int.

    the output image depth (-1 to use src.depth())

  • ksize: Size.

    kernel size

Keyword Arguments
  • anchor: Point.

    kernel anchor point. The default value of Point(-1, -1) denotes that the anchor is at the kernel center.

  • normalize: bool.

    flag, specifying whether the kernel is to be normalized by it's area or not.

  • borderType: int.

    border mode used to extrapolate pixels outside of the image, see #BorderTypes. #BORDER_WRAP is not supported.

Return
  • dst: Evision.Mat.t().

    output image of the same size and type as src

For every pixel \f$ (x, y) \f$ in the source image, the function calculates the sum of squares of those neighboring pixel values which overlap the filter placed over the pixel \f$ (x, y) \f$. The unnormalized square box filter can be useful in computing local image statistics such as the local variance and standard deviation around the neighborhood of a pixel. @sa boxFilter

Python prototype (for reference only):

sqrBoxFilter(src, ddepth, ksize[, dst[, anchor[, normalize[, borderType]]]]) -> dst
@spec sqrt(Evision.Mat.maybe_mat_in()) :: Evision.Mat.t() | {:error, String.t()}

Calculates a square root of array elements.

Positional Arguments
  • src: Evision.Mat.t().

    input floating-point array.

Return
  • dst: Evision.Mat.t().

    output array of the same size and type as src.

The function cv::sqrt calculates a square root of each input array element. In case of multi-channel arrays, each channel is processed independently. The accuracy is approximately the same as of the built-in std::sqrt .

Python prototype (for reference only):

sqrt(src[, dst]) -> dst
@spec sqrt(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Calculates a square root of array elements.

Positional Arguments
  • src: Evision.Mat.t().

    input floating-point array.

Return
  • dst: Evision.Mat.t().

    output array of the same size and type as src.

The function cv::sqrt calculates a square root of each input array element. In case of multi-channel arrays, each channel is processed independently. The accuracy is approximately the same as of the built-in std::sqrt .

Python prototype (for reference only):

sqrt(src[, dst]) -> dst
@spec stackBlur(
  Evision.Mat.maybe_mat_in(),
  {number(), number()}
) :: Evision.Mat.t() | {:error, String.t()}

Blurs an image using the stackBlur.

Positional Arguments
  • src: Evision.Mat.t().

    input image. The number of channels can be arbitrary, but the depth should be one of CV_8U, CV_16U, CV_16S or CV_32F.

  • ksize: Size.

    stack-blurring kernel size. The ksize.width and ksize.height can differ but they both must be positive and odd.

Return
  • dst: Evision.Mat.t().

    output image of the same size and type as src.

The function applies and stackBlur to an image. stackBlur can generate similar results as Gaussian blur, and the time consumption does not increase with the increase of kernel size. It creates a kind of moving stack of colors whilst scanning through the image. Thereby it just has to add one new block of color to the right side of the stack and remove the leftmost color. The remaining colors on the topmost layer of the stack are either added on or reduced by one, depending on if they are on the right or on the left side of the stack. The only supported borderType is BORDER_REPLICATE. Original paper was proposed by Mario Klingemann, which can be found http://underdestruction.com/2004/02/25/stackblur-2004.

Python prototype (for reference only):

stackBlur(src, ksize[, dst]) -> dst
Link to this function

stackBlur(src, ksize, opts)

View Source
@spec stackBlur(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Blurs an image using the stackBlur.

Positional Arguments
  • src: Evision.Mat.t().

    input image. The number of channels can be arbitrary, but the depth should be one of CV_8U, CV_16U, CV_16S or CV_32F.

  • ksize: Size.

    stack-blurring kernel size. The ksize.width and ksize.height can differ but they both must be positive and odd.

Return
  • dst: Evision.Mat.t().

    output image of the same size and type as src.

The function applies and stackBlur to an image. stackBlur can generate similar results as Gaussian blur, and the time consumption does not increase with the increase of kernel size. It creates a kind of moving stack of colors whilst scanning through the image. Thereby it just has to add one new block of color to the right side of the stack and remove the leftmost color. The remaining colors on the topmost layer of the stack are either added on or reduced by one, depending on if they are on the right or on the left side of the stack. The only supported borderType is BORDER_REPLICATE. Original paper was proposed by Mario Klingemann, which can be found http://underdestruction.com/2004/02/25/stackblur-2004.

Python prototype (for reference only):

stackBlur(src, ksize[, dst]) -> dst
@spec startWindowThread() :: integer() | {:error, String.t()}

startWindowThread

Return
  • retval: int

Python prototype (for reference only):

startWindowThread() -> retval
Link to this function

stereoCalibrate(objectPoints, imagePoints1, imagePoints2, cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2, imageSize)

View Source

stereoCalibrate

Positional Arguments
  • objectPoints: [Evision.Mat]
  • imagePoints1: [Evision.Mat]
  • imagePoints2: [Evision.Mat]
  • imageSize: Size
Keyword Arguments
  • flags: int.
  • criteria: TermCriteria.
Return
  • retval: double
  • cameraMatrix1: Evision.Mat.t()
  • distCoeffs1: Evision.Mat.t()
  • cameraMatrix2: Evision.Mat.t()
  • distCoeffs2: Evision.Mat.t()
  • r: Evision.Mat.t().
  • t: Evision.Mat.t().
  • e: Evision.Mat.t().
  • f: Evision.Mat.t().

Python prototype (for reference only):

stereoCalibrate(objectPoints, imagePoints1, imagePoints2, cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2, imageSize[, R[, T[, E[, F[, flags[, criteria]]]]]]) -> retval, cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2, R, T, E, F
Link to this function

stereoCalibrate(objectPoints, imagePoints1, imagePoints2, cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2, imageSize, opts)

View Source

stereoCalibrate

Positional Arguments
  • objectPoints: [Evision.Mat]
  • imagePoints1: [Evision.Mat]
  • imagePoints2: [Evision.Mat]
  • imageSize: Size
Keyword Arguments
  • flags: int.
  • criteria: TermCriteria.
Return
  • retval: double
  • cameraMatrix1: Evision.Mat.t()
  • distCoeffs1: Evision.Mat.t()
  • cameraMatrix2: Evision.Mat.t()
  • distCoeffs2: Evision.Mat.t()
  • r: Evision.Mat.t().
  • t: Evision.Mat.t().
  • e: Evision.Mat.t().
  • f: Evision.Mat.t().

Python prototype (for reference only):

stereoCalibrate(objectPoints, imagePoints1, imagePoints2, cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2, imageSize[, R[, T[, E[, F[, flags[, criteria]]]]]]) -> retval, cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2, R, T, E, F
Link to this function

stereoCalibrate(objectPoints, imagePoints1, imagePoints2, cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2, imageSize, r, t)

View Source

stereoCalibrate

Positional Arguments
  • objectPoints: [Evision.Mat]
  • imagePoints1: [Evision.Mat]
  • imagePoints2: [Evision.Mat]
  • imageSize: Size
Keyword Arguments
  • flags: int.
  • criteria: TermCriteria.
Return
  • retval: double
  • cameraMatrix1: Evision.Mat.t()
  • distCoeffs1: Evision.Mat.t()
  • cameraMatrix2: Evision.Mat.t()
  • distCoeffs2: Evision.Mat.t()
  • r: Evision.Mat.t()
  • t: Evision.Mat.t()
  • e: Evision.Mat.t().
  • f: Evision.Mat.t().
  • perViewErrors: Evision.Mat.t().

Python prototype (for reference only):

stereoCalibrate(objectPoints, imagePoints1, imagePoints2, cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2, imageSize, R, T[, E[, F[, perViewErrors[, flags[, criteria]]]]]) -> retval, cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2, R, T, E, F, perViewErrors
Link to this function

stereoCalibrate(objectPoints, imagePoints1, imagePoints2, cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2, imageSize, r, t, opts)

View Source

stereoCalibrate

Positional Arguments
  • objectPoints: [Evision.Mat]
  • imagePoints1: [Evision.Mat]
  • imagePoints2: [Evision.Mat]
  • imageSize: Size
Keyword Arguments
  • flags: int.
  • criteria: TermCriteria.
Return
  • retval: double
  • cameraMatrix1: Evision.Mat.t()
  • distCoeffs1: Evision.Mat.t()
  • cameraMatrix2: Evision.Mat.t()
  • distCoeffs2: Evision.Mat.t()
  • r: Evision.Mat.t()
  • t: Evision.Mat.t()
  • e: Evision.Mat.t().
  • f: Evision.Mat.t().
  • perViewErrors: Evision.Mat.t().

Python prototype (for reference only):

stereoCalibrate(objectPoints, imagePoints1, imagePoints2, cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2, imageSize, R, T[, E[, F[, perViewErrors[, flags[, criteria]]]]]) -> retval, cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2, R, T, E, F, perViewErrors
Link to this function

stereoCalibrateExtended(objectPoints, imagePoints1, imagePoints2, cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2, imageSize, r, t)

View Source

Calibrates a stereo camera set up. This function finds the intrinsic parameters for each of the two cameras and the extrinsic parameters between the two cameras.

Positional Arguments
  • objectPoints: [Evision.Mat].

    Vector of vectors of the calibration pattern points. The same structure as in @ref calibrateCamera. For each pattern view, both cameras need to see the same object points. Therefore, objectPoints.size(), imagePoints1.size(), and imagePoints2.size() need to be equal as well as objectPoints[i].size(), imagePoints1[i].size(), and imagePoints2[i].size() need to be equal for each i.

  • imagePoints1: [Evision.Mat].

    Vector of vectors of the projections of the calibration pattern points, observed by the first camera. The same structure as in @ref calibrateCamera.

  • imagePoints2: [Evision.Mat].

    Vector of vectors of the projections of the calibration pattern points, observed by the second camera. The same structure as in @ref calibrateCamera.

  • imageSize: Size.

    Size of the image used only to initialize the camera intrinsic matrices.

Keyword Arguments
  • flags: int.

    Different flags that may be zero or a combination of the following values:

    • @ref CALIB_FIX_INTRINSIC Fix cameraMatrix? and distCoeffs? so that only R, T, E, and F matrices are estimated.
    • @ref CALIB_USE_INTRINSIC_GUESS Optimize some or all of the intrinsic parameters according to the specified flags. Initial values are provided by the user.
    • @ref CALIB_USE_EXTRINSIC_GUESS R and T contain valid initial values that are optimized further. Otherwise R and T are initialized to the median value of the pattern views (each dimension separately).
    • @ref CALIB_FIX_PRINCIPAL_POINT Fix the principal points during the optimization.
    • @ref CALIB_FIX_FOCAL_LENGTH Fix \f$f^{(j)}_x\f$ and \f$f^{(j)}_y\f$ .
    • @ref CALIB_FIX_ASPECT_RATIO Optimize \f$f^{(j)}_y\f$ . Fix the ratio \f$f^{(j)}_x/f^{(j)}_y\f$ .
    • @ref CALIB_SAME_FOCAL_LENGTH Enforce \f$f^{(0)}_x=f^{(1)}_x\f$ and \f$f^{(0)}_y=f^{(1)}_y\f$ .
    • @ref CALIB_ZERO_TANGENT_DIST Set tangential distortion coefficients for each camera to zeros and fix there.
    • @ref CALIB_FIX_K1,..., @ref CALIB_FIX_K6 Do not change the corresponding radial distortion coefficient during the optimization. If @ref CALIB_USE_INTRINSIC_GUESS is set, the coefficient from the supplied distCoeffs matrix is used. Otherwise, it is set to 0.
    • @ref CALIB_RATIONAL_MODEL Enable coefficients k4, k5, and k6. To provide the backward compatibility, this extra flag should be explicitly specified to make the calibration function use the rational model and return 8 coefficients. If the flag is not set, the function computes and returns only 5 distortion coefficients.
    • @ref CALIB_THIN_PRISM_MODEL Coefficients s1, s2, s3 and s4 are enabled. To provide the backward compatibility, this extra flag should be explicitly specified to make the calibration function use the thin prism model and return 12 coefficients. If the flag is not set, the function computes and returns only 5 distortion coefficients.
    • @ref CALIB_FIX_S1_S2_S3_S4 The thin prism distortion coefficients are not changed during the optimization. If @ref CALIB_USE_INTRINSIC_GUESS is set, the coefficient from the supplied distCoeffs matrix is used. Otherwise, it is set to 0.
    • @ref CALIB_TILTED_MODEL Coefficients tauX and tauY are enabled. To provide the backward compatibility, this extra flag should be explicitly specified to make the calibration function use the tilted sensor model and return 14 coefficients. If the flag is not set, the function computes and returns only 5 distortion coefficients.
    • @ref CALIB_FIX_TAUX_TAUY The coefficients of the tilted sensor model are not changed during the optimization. If @ref CALIB_USE_INTRINSIC_GUESS is set, the coefficient from the supplied distCoeffs matrix is used. Otherwise, it is set to 0.
  • criteria: TermCriteria.

    Termination criteria for the iterative optimization algorithm.

Return
  • retval: double

  • cameraMatrix1: Evision.Mat.t().

    Input/output camera intrinsic matrix for the first camera, the same as in

  • distCoeffs1: Evision.Mat.t().

    Input/output vector of distortion coefficients, the same as in

  • cameraMatrix2: Evision.Mat.t().

    Input/output second camera intrinsic matrix for the second camera. See description for cameraMatrix1.

  • distCoeffs2: Evision.Mat.t().

    Input/output lens distortion coefficients for the second camera. See description for distCoeffs1.

  • r: Evision.Mat.t().

    Output rotation matrix. Together with the translation vector T, this matrix brings points given in the first camera's coordinate system to points in the second camera's coordinate system. In more technical terms, the tuple of R and T performs a change of basis from the first camera's coordinate system to the second camera's coordinate system. Due to its duality, this tuple is equivalent to the position of the first camera with respect to the second camera coordinate system.

  • t: Evision.Mat.t().

    Output translation vector, see description above.

  • e: Evision.Mat.t().

    Output essential matrix.

  • f: Evision.Mat.t().

    Output fundamental matrix.

  • rvecs: [Evision.Mat].

    Output vector of rotation vectors ( @ref Rodrigues ) estimated for each pattern view in the coordinate system of the first camera of the stereo pair (e.g. std::vector<cv::Mat>). More in detail, each i-th rotation vector together with the corresponding i-th translation vector (see the next output parameter description) brings the calibration pattern from the object coordinate space (in which object points are specified) to the camera coordinate space of the first camera of the stereo pair. In more technical terms, the tuple of the i-th rotation and translation vector performs a change of basis from object coordinate space to camera coordinate space of the first camera of the stereo pair.

  • tvecs: [Evision.Mat].

    Output vector of translation vectors estimated for each pattern view, see parameter description of previous output parameter ( rvecs ).

  • perViewErrors: Evision.Mat.t().

    Output vector of the RMS re-projection error estimated for each pattern view.

@ref calibrateCamera. Furthermore, for the stereo case, additional flags may be used, see below. @ref calibrateCamera.

The function estimates the transformation between two cameras making a stereo pair. If one computes the poses of an object relative to the first camera and to the second camera, ( \f$R_1\f$,\f$T_1\f$ ) and (\f$R_2\f$,\f$T_2\f$), respectively, for a stereo camera where the relative position and orientation between the two cameras are fixed, then those poses definitely relate to each other. This means, if the relative position and orientation (\f$R\f$,\f$T\f$) of the two cameras is known, it is possible to compute (\f$R_2\f$,\f$T_2\f$) when (\f$R_1\f$,\f$T_1\f$) is given. This is what the described function does. It computes (\f$R\f$,\f$T\f$) such that: \f[R_2=R R_1\f] \f[T_2=R T_1 + T.\f] Therefore, one can compute the coordinate representation of a 3D point for the second camera's coordinate system when given the point's coordinate representation in the first camera's coordinate system: \f[\begin{bmatrix} X_2 \\ Y_2 \\ Z_2 \\ 1 \end{bmatrix} = \begin{bmatrix} R & T \\ 0 & 1 \end{bmatrix} \begin{bmatrix} X_1 \\ Y_1 \\ Z_1 \\ 1 \end{bmatrix}.\f] Optionally, it computes the essential matrix E: \f[E= \vecthreethree{0}{-T_2}{T_1}{T_2}{0}{-T_0}{-T_1}{T_0}{0} R\f] where \f$T_i\f$ are components of the translation vector \f$T\f$ : \f$T=[T_0, T_1, T_2]^T\f$ . And the function can also compute the fundamental matrix F: \f[F = cameraMatrix2^{-T}\cdot E \cdot cameraMatrix1^{-1}\f] Besides the stereo-related information, the function can also perform a full calibration of each of the two cameras. However, due to the high dimensionality of the parameter space and noise in the input data, the function can diverge from the correct solution. If the intrinsic parameters can be estimated with high accuracy for each of the cameras individually (for example, using #calibrateCamera ), you are recommended to do so and then pass @ref CALIB_FIX_INTRINSIC flag to the function along with the computed intrinsic parameters. Otherwise, if all the parameters are estimated at once, it makes sense to restrict some parameters, for example, pass @ref CALIB_SAME_FOCAL_LENGTH and @ref CALIB_ZERO_TANGENT_DIST flags, which is usually a reasonable assumption. Similarly to #calibrateCamera, the function minimizes the total re-projection error for all the points in all the available views from both cameras. The function returns the final value of the re-projection error.

Python prototype (for reference only):

stereoCalibrateExtended(objectPoints, imagePoints1, imagePoints2, cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2, imageSize, R, T[, E[, F[, rvecs[, tvecs[, perViewErrors[, flags[, criteria]]]]]]]) -> retval, cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2, R, T, E, F, rvecs, tvecs, perViewErrors
Link to this function

stereoCalibrateExtended(objectPoints, imagePoints1, imagePoints2, cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2, imageSize, r, t, opts)

View Source

Calibrates a stereo camera set up. This function finds the intrinsic parameters for each of the two cameras and the extrinsic parameters between the two cameras.

Positional Arguments
  • objectPoints: [Evision.Mat].

    Vector of vectors of the calibration pattern points. The same structure as in @ref calibrateCamera. For each pattern view, both cameras need to see the same object points. Therefore, objectPoints.size(), imagePoints1.size(), and imagePoints2.size() need to be equal as well as objectPoints[i].size(), imagePoints1[i].size(), and imagePoints2[i].size() need to be equal for each i.

  • imagePoints1: [Evision.Mat].

    Vector of vectors of the projections of the calibration pattern points, observed by the first camera. The same structure as in @ref calibrateCamera.

  • imagePoints2: [Evision.Mat].

    Vector of vectors of the projections of the calibration pattern points, observed by the second camera. The same structure as in @ref calibrateCamera.

  • imageSize: Size.

    Size of the image used only to initialize the camera intrinsic matrices.

Keyword Arguments
  • flags: int.

    Different flags that may be zero or a combination of the following values:

    • @ref CALIB_FIX_INTRINSIC Fix cameraMatrix? and distCoeffs? so that only R, T, E, and F matrices are estimated.
    • @ref CALIB_USE_INTRINSIC_GUESS Optimize some or all of the intrinsic parameters according to the specified flags. Initial values are provided by the user.
    • @ref CALIB_USE_EXTRINSIC_GUESS R and T contain valid initial values that are optimized further. Otherwise R and T are initialized to the median value of the pattern views (each dimension separately).
    • @ref CALIB_FIX_PRINCIPAL_POINT Fix the principal points during the optimization.
    • @ref CALIB_FIX_FOCAL_LENGTH Fix \f$f^{(j)}_x\f$ and \f$f^{(j)}_y\f$ .
    • @ref CALIB_FIX_ASPECT_RATIO Optimize \f$f^{(j)}_y\f$ . Fix the ratio \f$f^{(j)}_x/f^{(j)}_y\f$ .
    • @ref CALIB_SAME_FOCAL_LENGTH Enforce \f$f^{(0)}_x=f^{(1)}_x\f$ and \f$f^{(0)}_y=f^{(1)}_y\f$ .
    • @ref CALIB_ZERO_TANGENT_DIST Set tangential distortion coefficients for each camera to zeros and fix there.
    • @ref CALIB_FIX_K1,..., @ref CALIB_FIX_K6 Do not change the corresponding radial distortion coefficient during the optimization. If @ref CALIB_USE_INTRINSIC_GUESS is set, the coefficient from the supplied distCoeffs matrix is used. Otherwise, it is set to 0.
    • @ref CALIB_RATIONAL_MODEL Enable coefficients k4, k5, and k6. To provide the backward compatibility, this extra flag should be explicitly specified to make the calibration function use the rational model and return 8 coefficients. If the flag is not set, the function computes and returns only 5 distortion coefficients.
    • @ref CALIB_THIN_PRISM_MODEL Coefficients s1, s2, s3 and s4 are enabled. To provide the backward compatibility, this extra flag should be explicitly specified to make the calibration function use the thin prism model and return 12 coefficients. If the flag is not set, the function computes and returns only 5 distortion coefficients.
    • @ref CALIB_FIX_S1_S2_S3_S4 The thin prism distortion coefficients are not changed during the optimization. If @ref CALIB_USE_INTRINSIC_GUESS is set, the coefficient from the supplied distCoeffs matrix is used. Otherwise, it is set to 0.
    • @ref CALIB_TILTED_MODEL Coefficients tauX and tauY are enabled. To provide the backward compatibility, this extra flag should be explicitly specified to make the calibration function use the tilted sensor model and return 14 coefficients. If the flag is not set, the function computes and returns only 5 distortion coefficients.
    • @ref CALIB_FIX_TAUX_TAUY The coefficients of the tilted sensor model are not changed during the optimization. If @ref CALIB_USE_INTRINSIC_GUESS is set, the coefficient from the supplied distCoeffs matrix is used. Otherwise, it is set to 0.
  • criteria: TermCriteria.

    Termination criteria for the iterative optimization algorithm.

Return
  • retval: double

  • cameraMatrix1: Evision.Mat.t().

    Input/output camera intrinsic matrix for the first camera, the same as in

  • distCoeffs1: Evision.Mat.t().

    Input/output vector of distortion coefficients, the same as in

  • cameraMatrix2: Evision.Mat.t().

    Input/output second camera intrinsic matrix for the second camera. See description for cameraMatrix1.

  • distCoeffs2: Evision.Mat.t().

    Input/output lens distortion coefficients for the second camera. See description for distCoeffs1.

  • r: Evision.Mat.t().

    Output rotation matrix. Together with the translation vector T, this matrix brings points given in the first camera's coordinate system to points in the second camera's coordinate system. In more technical terms, the tuple of R and T performs a change of basis from the first camera's coordinate system to the second camera's coordinate system. Due to its duality, this tuple is equivalent to the position of the first camera with respect to the second camera coordinate system.

  • t: Evision.Mat.t().

    Output translation vector, see description above.

  • e: Evision.Mat.t().

    Output essential matrix.

  • f: Evision.Mat.t().

    Output fundamental matrix.

  • rvecs: [Evision.Mat].

    Output vector of rotation vectors ( @ref Rodrigues ) estimated for each pattern view in the coordinate system of the first camera of the stereo pair (e.g. std::vector<cv::Mat>). More in detail, each i-th rotation vector together with the corresponding i-th translation vector (see the next output parameter description) brings the calibration pattern from the object coordinate space (in which object points are specified) to the camera coordinate space of the first camera of the stereo pair. In more technical terms, the tuple of the i-th rotation and translation vector performs a change of basis from object coordinate space to camera coordinate space of the first camera of the stereo pair.

  • tvecs: [Evision.Mat].

    Output vector of translation vectors estimated for each pattern view, see parameter description of previous output parameter ( rvecs ).

  • perViewErrors: Evision.Mat.t().

    Output vector of the RMS re-projection error estimated for each pattern view.

@ref calibrateCamera. Furthermore, for the stereo case, additional flags may be used, see below. @ref calibrateCamera.

The function estimates the transformation between two cameras making a stereo pair. If one computes the poses of an object relative to the first camera and to the second camera, ( \f$R_1\f$,\f$T_1\f$ ) and (\f$R_2\f$,\f$T_2\f$), respectively, for a stereo camera where the relative position and orientation between the two cameras are fixed, then those poses definitely relate to each other. This means, if the relative position and orientation (\f$R\f$,\f$T\f$) of the two cameras is known, it is possible to compute (\f$R_2\f$,\f$T_2\f$) when (\f$R_1\f$,\f$T_1\f$) is given. This is what the described function does. It computes (\f$R\f$,\f$T\f$) such that: \f[R_2=R R_1\f] \f[T_2=R T_1 + T.\f] Therefore, one can compute the coordinate representation of a 3D point for the second camera's coordinate system when given the point's coordinate representation in the first camera's coordinate system: \f[\begin{bmatrix} X_2 \\ Y_2 \\ Z_2 \\ 1 \end{bmatrix} = \begin{bmatrix} R & T \\ 0 & 1 \end{bmatrix} \begin{bmatrix} X_1 \\ Y_1 \\ Z_1 \\ 1 \end{bmatrix}.\f] Optionally, it computes the essential matrix E: \f[E= \vecthreethree{0}{-T_2}{T_1}{T_2}{0}{-T_0}{-T_1}{T_0}{0} R\f] where \f$T_i\f$ are components of the translation vector \f$T\f$ : \f$T=[T_0, T_1, T_2]^T\f$ . And the function can also compute the fundamental matrix F: \f[F = cameraMatrix2^{-T}\cdot E \cdot cameraMatrix1^{-1}\f] Besides the stereo-related information, the function can also perform a full calibration of each of the two cameras. However, due to the high dimensionality of the parameter space and noise in the input data, the function can diverge from the correct solution. If the intrinsic parameters can be estimated with high accuracy for each of the cameras individually (for example, using #calibrateCamera ), you are recommended to do so and then pass @ref CALIB_FIX_INTRINSIC flag to the function along with the computed intrinsic parameters. Otherwise, if all the parameters are estimated at once, it makes sense to restrict some parameters, for example, pass @ref CALIB_SAME_FOCAL_LENGTH and @ref CALIB_ZERO_TANGENT_DIST flags, which is usually a reasonable assumption. Similarly to #calibrateCamera, the function minimizes the total re-projection error for all the points in all the available views from both cameras. The function returns the final value of the re-projection error.

Python prototype (for reference only):

stereoCalibrateExtended(objectPoints, imagePoints1, imagePoints2, cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2, imageSize, R, T[, E[, F[, rvecs[, tvecs[, perViewErrors[, flags[, criteria]]]]]]]) -> retval, cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2, R, T, E, F, rvecs, tvecs, perViewErrors
Link to this function

stereoRectify(cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2, imageSize, r, t)

View Source

Computes rectification transforms for each head of a calibrated stereo camera.

Positional Arguments
  • cameraMatrix1: Evision.Mat.t().

    First camera intrinsic matrix.

  • distCoeffs1: Evision.Mat.t().

    First camera distortion parameters.

  • cameraMatrix2: Evision.Mat.t().

    Second camera intrinsic matrix.

  • distCoeffs2: Evision.Mat.t().

    Second camera distortion parameters.

  • imageSize: Size.

    Size of the image used for stereo calibration.

  • r: Evision.Mat.t().

    Rotation matrix from the coordinate system of the first camera to the second camera, see @ref stereoCalibrate.

  • t: Evision.Mat.t().

    Translation vector from the coordinate system of the first camera to the second camera, see @ref stereoCalibrate.

Keyword Arguments
  • flags: int.

    Operation flags that may be zero or @ref CALIB_ZERO_DISPARITY . If the flag is set, the function makes the principal points of each camera have the same pixel coordinates in the rectified views. And if the flag is not set, the function may still shift the images in the horizontal or vertical direction (depending on the orientation of epipolar lines) to maximize the useful image area.

  • alpha: double.

    Free scaling parameter. If it is -1 or absent, the function performs the default scaling. Otherwise, the parameter should be between 0 and 1. alpha=0 means that the rectified images are zoomed and shifted so that only valid pixels are visible (no black areas after rectification). alpha=1 means that the rectified image is decimated and shifted so that all the pixels from the original images from the cameras are retained in the rectified images (no source image pixels are lost). Any intermediate value yields an intermediate result between those two extreme cases.

  • newImageSize: Size.

    New image resolution after rectification. The same size should be passed to #initUndistortRectifyMap (see the stereo_calib.cpp sample in OpenCV samples directory). When (0,0) is passed (default), it is set to the original imageSize . Setting it to a larger value can help you preserve details in the original image, especially when there is a big radial distortion.

Return
  • r1: Evision.Mat.t().

    Output 3x3 rectification transform (rotation matrix) for the first camera. This matrix brings points given in the unrectified first camera's coordinate system to points in the rectified first camera's coordinate system. In more technical terms, it performs a change of basis from the unrectified first camera's coordinate system to the rectified first camera's coordinate system.

  • r2: Evision.Mat.t().

    Output 3x3 rectification transform (rotation matrix) for the second camera. This matrix brings points given in the unrectified second camera's coordinate system to points in the rectified second camera's coordinate system. In more technical terms, it performs a change of basis from the unrectified second camera's coordinate system to the rectified second camera's coordinate system.

  • p1: Evision.Mat.t().

    Output 3x4 projection matrix in the new (rectified) coordinate systems for the first camera, i.e. it projects points given in the rectified first camera coordinate system into the rectified first camera's image.

  • p2: Evision.Mat.t().

    Output 3x4 projection matrix in the new (rectified) coordinate systems for the second camera, i.e. it projects points given in the rectified first camera coordinate system into the rectified second camera's image.

  • q: Evision.Mat.t().

    Output \f$4 \times 4\f$ disparity-to-depth mapping matrix (see @ref reprojectImageTo3D).

  • validPixROI1: Rect*.

    Optional output rectangles inside the rectified images where all the pixels are valid. If alpha=0 , the ROIs cover the whole images. Otherwise, they are likely to be smaller (see the picture below).

  • validPixROI2: Rect*.

    Optional output rectangles inside the rectified images where all the pixels are valid. If alpha=0 , the ROIs cover the whole images. Otherwise, they are likely to be smaller (see the picture below).

The function computes the rotation matrices for each camera that (virtually) make both camera image planes the same plane. Consequently, this makes all the epipolar lines parallel and thus simplifies the dense stereo correspondence problem. The function takes the matrices computed by #stereoCalibrate as input. As output, it provides two rotation matrices and also two projection matrices in the new coordinates. The function distinguishes the following two cases:

  • Horizontal stereo: the first and the second camera views are shifted relative to each other mainly along the x-axis (with possible small vertical shift). In the rectified images, the corresponding epipolar lines in the left and right cameras are horizontal and have the same y-coordinate. P1 and P2 look like:

\f[\texttt{P1} = \begin{bmatrix} f & 0 & cx_1 & 0 \\ 0 & f & cy & 0 \\ 0 & 0 & 1 & 0 \end{bmatrix}\f] \f[\texttt{P2} = \begin{bmatrix} f & 0 & cx_2 & T_x \cdot f \\ 0 & f & cy & 0 \\ 0 & 0 & 1 & 0 \end{bmatrix} ,\f] \f[\texttt{Q} = \begin{bmatrix} 1 & 0 & 0 & -cx_1 \\ 0 & 1 & 0 & -cy \\ 0 & 0 & 0 & f \\ 0 & 0 & -\frac{1}{T_x} & \frac{cx_1 - cx_2}{T_x} \end{bmatrix} \f] where \f$T_x\f$ is a horizontal shift between the cameras and \f$cx_1=cx_2\f$ if @ref CALIB_ZERO_DISPARITY is set.

  • Vertical stereo: the first and the second camera views are shifted relative to each other mainly in the vertical direction (and probably a bit in the horizontal direction too). The epipolar lines in the rectified images are vertical and have the same x-coordinate. P1 and P2 look like:

\f[\texttt{P1} = \begin{bmatrix} f & 0 & cx & 0 \\ 0 & f & cy_1 & 0 \\ 0 & 0 & 1 & 0 \end{bmatrix}\f] \f[\texttt{P2} = \begin{bmatrix} f & 0 & cx & 0 \\ 0 & f & cy_2 & T_y \cdot f \\ 0 & 0 & 1 & 0 \end{bmatrix},\f] \f[\texttt{Q} = \begin{bmatrix} 1 & 0 & 0 & -cx \\ 0 & 1 & 0 & -cy_1 \\ 0 & 0 & 0 & f \\ 0 & 0 & -\frac{1}{T_y} & \frac{cy_1 - cy_2}{T_y} \end{bmatrix} \f] where \f$T_y\f$ is a vertical shift between the cameras and \f$cy_1=cy_2\f$ if @ref CALIB_ZERO_DISPARITY is set. As you can see, the first three columns of P1 and P2 will effectively be the new "rectified" camera matrices. The matrices, together with R1 and R2 , can then be passed to #initUndistortRectifyMap to initialize the rectification map for each camera. See below the screenshot from the stereo_calib.cpp sample. Some red horizontal lines pass through the corresponding image regions. This means that the images are well rectified, which is what most stereo correspondence algorithms rely on. The green rectangles are roi1 and roi2 . You see that their interiors are all valid pixels. image

Python prototype (for reference only):

stereoRectify(cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2, imageSize, R, T[, R1[, R2[, P1[, P2[, Q[, flags[, alpha[, newImageSize]]]]]]]]) -> R1, R2, P1, P2, Q, validPixROI1, validPixROI2
Link to this function

stereoRectify(cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2, imageSize, r, t, opts)

View Source

Computes rectification transforms for each head of a calibrated stereo camera.

Positional Arguments
  • cameraMatrix1: Evision.Mat.t().

    First camera intrinsic matrix.

  • distCoeffs1: Evision.Mat.t().

    First camera distortion parameters.

  • cameraMatrix2: Evision.Mat.t().

    Second camera intrinsic matrix.

  • distCoeffs2: Evision.Mat.t().

    Second camera distortion parameters.

  • imageSize: Size.

    Size of the image used for stereo calibration.

  • r: Evision.Mat.t().

    Rotation matrix from the coordinate system of the first camera to the second camera, see @ref stereoCalibrate.

  • t: Evision.Mat.t().

    Translation vector from the coordinate system of the first camera to the second camera, see @ref stereoCalibrate.

Keyword Arguments
  • flags: int.

    Operation flags that may be zero or @ref CALIB_ZERO_DISPARITY . If the flag is set, the function makes the principal points of each camera have the same pixel coordinates in the rectified views. And if the flag is not set, the function may still shift the images in the horizontal or vertical direction (depending on the orientation of epipolar lines) to maximize the useful image area.

  • alpha: double.

    Free scaling parameter. If it is -1 or absent, the function performs the default scaling. Otherwise, the parameter should be between 0 and 1. alpha=0 means that the rectified images are zoomed and shifted so that only valid pixels are visible (no black areas after rectification). alpha=1 means that the rectified image is decimated and shifted so that all the pixels from the original images from the cameras are retained in the rectified images (no source image pixels are lost). Any intermediate value yields an intermediate result between those two extreme cases.

  • newImageSize: Size.

    New image resolution after rectification. The same size should be passed to #initUndistortRectifyMap (see the stereo_calib.cpp sample in OpenCV samples directory). When (0,0) is passed (default), it is set to the original imageSize . Setting it to a larger value can help you preserve details in the original image, especially when there is a big radial distortion.

Return
  • r1: Evision.Mat.t().

    Output 3x3 rectification transform (rotation matrix) for the first camera. This matrix brings points given in the unrectified first camera's coordinate system to points in the rectified first camera's coordinate system. In more technical terms, it performs a change of basis from the unrectified first camera's coordinate system to the rectified first camera's coordinate system.

  • r2: Evision.Mat.t().

    Output 3x3 rectification transform (rotation matrix) for the second camera. This matrix brings points given in the unrectified second camera's coordinate system to points in the rectified second camera's coordinate system. In more technical terms, it performs a change of basis from the unrectified second camera's coordinate system to the rectified second camera's coordinate system.

  • p1: Evision.Mat.t().

    Output 3x4 projection matrix in the new (rectified) coordinate systems for the first camera, i.e. it projects points given in the rectified first camera coordinate system into the rectified first camera's image.

  • p2: Evision.Mat.t().

    Output 3x4 projection matrix in the new (rectified) coordinate systems for the second camera, i.e. it projects points given in the rectified first camera coordinate system into the rectified second camera's image.

  • q: Evision.Mat.t().

    Output \f$4 \times 4\f$ disparity-to-depth mapping matrix (see @ref reprojectImageTo3D).

  • validPixROI1: Rect*.

    Optional output rectangles inside the rectified images where all the pixels are valid. If alpha=0 , the ROIs cover the whole images. Otherwise, they are likely to be smaller (see the picture below).

  • validPixROI2: Rect*.

    Optional output rectangles inside the rectified images where all the pixels are valid. If alpha=0 , the ROIs cover the whole images. Otherwise, they are likely to be smaller (see the picture below).

The function computes the rotation matrices for each camera that (virtually) make both camera image planes the same plane. Consequently, this makes all the epipolar lines parallel and thus simplifies the dense stereo correspondence problem. The function takes the matrices computed by #stereoCalibrate as input. As output, it provides two rotation matrices and also two projection matrices in the new coordinates. The function distinguishes the following two cases:

  • Horizontal stereo: the first and the second camera views are shifted relative to each other mainly along the x-axis (with possible small vertical shift). In the rectified images, the corresponding epipolar lines in the left and right cameras are horizontal and have the same y-coordinate. P1 and P2 look like:

\f[\texttt{P1} = \begin{bmatrix} f & 0 & cx_1 & 0 \\ 0 & f & cy & 0 \\ 0 & 0 & 1 & 0 \end{bmatrix}\f] \f[\texttt{P2} = \begin{bmatrix} f & 0 & cx_2 & T_x \cdot f \\ 0 & f & cy & 0 \\ 0 & 0 & 1 & 0 \end{bmatrix} ,\f] \f[\texttt{Q} = \begin{bmatrix} 1 & 0 & 0 & -cx_1 \\ 0 & 1 & 0 & -cy \\ 0 & 0 & 0 & f \\ 0 & 0 & -\frac{1}{T_x} & \frac{cx_1 - cx_2}{T_x} \end{bmatrix} \f] where \f$T_x\f$ is a horizontal shift between the cameras and \f$cx_1=cx_2\f$ if @ref CALIB_ZERO_DISPARITY is set.

  • Vertical stereo: the first and the second camera views are shifted relative to each other mainly in the vertical direction (and probably a bit in the horizontal direction too). The epipolar lines in the rectified images are vertical and have the same x-coordinate. P1 and P2 look like:

\f[\texttt{P1} = \begin{bmatrix} f & 0 & cx & 0 \\ 0 & f & cy_1 & 0 \\ 0 & 0 & 1 & 0 \end{bmatrix}\f] \f[\texttt{P2} = \begin{bmatrix} f & 0 & cx & 0 \\ 0 & f & cy_2 & T_y \cdot f \\ 0 & 0 & 1 & 0 \end{bmatrix},\f] \f[\texttt{Q} = \begin{bmatrix} 1 & 0 & 0 & -cx \\ 0 & 1 & 0 & -cy_1 \\ 0 & 0 & 0 & f \\ 0 & 0 & -\frac{1}{T_y} & \frac{cy_1 - cy_2}{T_y} \end{bmatrix} \f] where \f$T_y\f$ is a vertical shift between the cameras and \f$cy_1=cy_2\f$ if @ref CALIB_ZERO_DISPARITY is set. As you can see, the first three columns of P1 and P2 will effectively be the new "rectified" camera matrices. The matrices, together with R1 and R2 , can then be passed to #initUndistortRectifyMap to initialize the rectification map for each camera. See below the screenshot from the stereo_calib.cpp sample. Some red horizontal lines pass through the corresponding image regions. This means that the images are well rectified, which is what most stereo correspondence algorithms rely on. The green rectangles are roi1 and roi2 . You see that their interiors are all valid pixels. image

Python prototype (for reference only):

stereoRectify(cameraMatrix1, distCoeffs1, cameraMatrix2, distCoeffs2, imageSize, R, T[, R1[, R2[, P1[, P2[, Q[, flags[, alpha[, newImageSize]]]]]]]]) -> R1, R2, P1, P2, Q, validPixROI1, validPixROI2
Link to this function

stereoRectifyUncalibrated(points1, points2, f, imgSize)

View Source
@spec stereoRectifyUncalibrated(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  {number(), number()}
) :: {Evision.Mat.t(), Evision.Mat.t()} | false | {:error, String.t()}

Computes a rectification transform for an uncalibrated stereo camera.

Positional Arguments
  • points1: Evision.Mat.t().

    Array of feature points in the first image.

  • points2: Evision.Mat.t().

    The corresponding points in the second image. The same formats as in #findFundamentalMat are supported.

  • f: Evision.Mat.t().

    Input fundamental matrix. It can be computed from the same set of point pairs using #findFundamentalMat .

  • imgSize: Size.

    Size of the image.

Keyword Arguments
  • threshold: double.

    Optional threshold used to filter out the outliers. If the parameter is greater than zero, all the point pairs that do not comply with the epipolar geometry (that is, the points for which \f$|\texttt{points2[i]}^T \cdot \texttt{F} \cdot \texttt{points1[i]}|>\texttt{threshold}\f$ ) are rejected prior to computing the homographies. Otherwise, all the points are considered inliers.

Return
  • retval: bool

  • h1: Evision.Mat.t().

    Output rectification homography matrix for the first image.

  • h2: Evision.Mat.t().

    Output rectification homography matrix for the second image.

The function computes the rectification transformations without knowing intrinsic parameters of the cameras and their relative position in the space, which explains the suffix "uncalibrated". Another related difference from #stereoRectify is that the function outputs not the rectification transformations in the object (3D) space, but the planar perspective transformations encoded by the homography matrices H1 and H2 . The function implements the algorithm @cite Hartley99 . Note: While the algorithm does not need to know the intrinsic parameters of the cameras, it heavily depends on the epipolar geometry. Therefore, if the camera lenses have a significant distortion, it would be better to correct it before computing the fundamental matrix and calling this function. For example, distortion coefficients can be estimated for each head of stereo camera separately by using #calibrateCamera . Then, the images can be corrected using #undistort , or just the point coordinates can be corrected with #undistortPoints .

Python prototype (for reference only):

stereoRectifyUncalibrated(points1, points2, F, imgSize[, H1[, H2[, threshold]]]) -> retval, H1, H2
Link to this function

stereoRectifyUncalibrated(points1, points2, f, imgSize, opts)

View Source
@spec stereoRectifyUncalibrated(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  [{atom(), term()}, ...] | nil
) :: {Evision.Mat.t(), Evision.Mat.t()} | false | {:error, String.t()}

Computes a rectification transform for an uncalibrated stereo camera.

Positional Arguments
  • points1: Evision.Mat.t().

    Array of feature points in the first image.

  • points2: Evision.Mat.t().

    The corresponding points in the second image. The same formats as in #findFundamentalMat are supported.

  • f: Evision.Mat.t().

    Input fundamental matrix. It can be computed from the same set of point pairs using #findFundamentalMat .

  • imgSize: Size.

    Size of the image.

Keyword Arguments
  • threshold: double.

    Optional threshold used to filter out the outliers. If the parameter is greater than zero, all the point pairs that do not comply with the epipolar geometry (that is, the points for which \f$|\texttt{points2[i]}^T \cdot \texttt{F} \cdot \texttt{points1[i]}|>\texttt{threshold}\f$ ) are rejected prior to computing the homographies. Otherwise, all the points are considered inliers.

Return
  • retval: bool

  • h1: Evision.Mat.t().

    Output rectification homography matrix for the first image.

  • h2: Evision.Mat.t().

    Output rectification homography matrix for the second image.

The function computes the rectification transformations without knowing intrinsic parameters of the cameras and their relative position in the space, which explains the suffix "uncalibrated". Another related difference from #stereoRectify is that the function outputs not the rectification transformations in the object (3D) space, but the planar perspective transformations encoded by the homography matrices H1 and H2 . The function implements the algorithm @cite Hartley99 . Note: While the algorithm does not need to know the intrinsic parameters of the cameras, it heavily depends on the epipolar geometry. Therefore, if the camera lenses have a significant distortion, it would be better to correct it before computing the fundamental matrix and calling this function. For example, distortion coefficients can be estimated for each head of stereo camera separately by using #calibrateCamera . Then, the images can be corrected using #undistort , or just the point coordinates can be corrected with #undistortPoints .

Python prototype (for reference only):

stereoRectifyUncalibrated(points1, points2, F, imgSize[, H1[, H2[, threshold]]]) -> retval, H1, H2
@spec stylization(Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

Stylization aims to produce digital imagery with a wide variety of effects not focused on photorealism. Edge-aware filters are ideal for stylization, as they can abstract regions of low contrast while preserving, or enhancing, high-contrast features.

Positional Arguments
  • src: Evision.Mat.t().

    Input 8-bit 3-channel image.

Keyword Arguments
  • sigma_s: float.

    %Range between 0 to 200.

  • sigma_r: float.

    %Range between 0 to 1.

Return
  • dst: Evision.Mat.t().

    Output image with the same size and type as src.

Python prototype (for reference only):

stylization(src[, dst[, sigma_s[, sigma_r]]]) -> dst
@spec stylization(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Stylization aims to produce digital imagery with a wide variety of effects not focused on photorealism. Edge-aware filters are ideal for stylization, as they can abstract regions of low contrast while preserving, or enhancing, high-contrast features.

Positional Arguments
  • src: Evision.Mat.t().

    Input 8-bit 3-channel image.

Keyword Arguments
  • sigma_s: float.

    %Range between 0 to 200.

  • sigma_r: float.

    %Range between 0 to 1.

Return
  • dst: Evision.Mat.t().

    Output image with the same size and type as src.

Python prototype (for reference only):

stylization(src[, dst[, sigma_s[, sigma_r]]]) -> dst

Calculates the per-element difference between two arrays or array and a scalar.

Positional Arguments
  • src1: Evision.Mat.t().

    first input array or a scalar.

  • src2: Evision.Mat.t().

    second input array or a scalar.

Keyword Arguments
  • mask: Evision.Mat.t().

    optional operation mask; this is an 8-bit single channel array that specifies elements of the output array to be changed.

  • dtype: int.

    optional depth of the output array

Return
  • dst: Evision.Mat.t().

    output array of the same size and the same number of channels as the input array.

The function subtract calculates:

  • Difference between two arrays, when both input arrays have the same size and the same number of channels: \f[\texttt{dst}(I) = \texttt{saturate} ( \texttt{src1}(I) - \texttt{src2}(I)) \quad \texttt{if mask}(I) \ne0\f]

  • Difference between an array and a scalar, when src2 is constructed from Scalar or has the same number of elements as src1.channels(): \f[\texttt{dst}(I) = \texttt{saturate} ( \texttt{src1}(I) - \texttt{src2} ) \quad \texttt{if mask}(I) \ne0\f]

  • Difference between a scalar and an array, when src1 is constructed from Scalar or has the same number of elements as src2.channels(): \f[\texttt{dst}(I) = \texttt{saturate} ( \texttt{src1} - \texttt{src2}(I) ) \quad \texttt{if mask}(I) \ne0\f]

  • The reverse difference between a scalar and an array in the case of SubRS: \f[\texttt{dst}(I) = \texttt{saturate} ( \texttt{src2} - \texttt{src1}(I) ) \quad \texttt{if mask}(I) \ne0\f] where I is a multi-dimensional index of array elements. In case of multi-channel arrays, each channel is processed independently.

The first function in the list above can be replaced with matrix expressions:

dst = src1 - src2;
dst -= src1; // equivalent to subtract(dst, src1, dst);

The input arrays and the output array can all have the same or different depths. For example, you can subtract to 8-bit unsigned arrays and store the difference in a 16-bit signed array. Depth of the output array is determined by dtype parameter. In the second and third cases above, as well as in the first case, when src1.depth() == src2.depth(), dtype can be set to the default -1. In this case the output array will have the same depth as the input array, be it src1, src2 or both. Note: Saturation is not applied when the output array has the depth CV_32S. You may even get result of an incorrect sign in the case of overflow. Note: (Python) Be careful to difference behaviour between src1/src2 are single number and they are tuple/array. subtract(src,X) means subtract(src,(X,X,X,X)). subtract(src,(X,)) means subtract(src,(X,0,0,0)). @sa add, addWeighted, scaleAdd, Mat::convertTo

Python prototype (for reference only):

subtract(src1, src2[, dst[, mask[, dtype]]]) -> dst
Link to this function

subtract(src1, src2, opts)

View Source
@spec subtract(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Calculates the per-element difference between two arrays or array and a scalar.

Positional Arguments
  • src1: Evision.Mat.t().

    first input array or a scalar.

  • src2: Evision.Mat.t().

    second input array or a scalar.

Keyword Arguments
  • mask: Evision.Mat.t().

    optional operation mask; this is an 8-bit single channel array that specifies elements of the output array to be changed.

  • dtype: int.

    optional depth of the output array

Return
  • dst: Evision.Mat.t().

    output array of the same size and the same number of channels as the input array.

The function subtract calculates:

  • Difference between two arrays, when both input arrays have the same size and the same number of channels: \f[\texttt{dst}(I) = \texttt{saturate} ( \texttt{src1}(I) - \texttt{src2}(I)) \quad \texttt{if mask}(I) \ne0\f]

  • Difference between an array and a scalar, when src2 is constructed from Scalar or has the same number of elements as src1.channels(): \f[\texttt{dst}(I) = \texttt{saturate} ( \texttt{src1}(I) - \texttt{src2} ) \quad \texttt{if mask}(I) \ne0\f]

  • Difference between a scalar and an array, when src1 is constructed from Scalar or has the same number of elements as src2.channels(): \f[\texttt{dst}(I) = \texttt{saturate} ( \texttt{src1} - \texttt{src2}(I) ) \quad \texttt{if mask}(I) \ne0\f]

  • The reverse difference between a scalar and an array in the case of SubRS: \f[\texttt{dst}(I) = \texttt{saturate} ( \texttt{src2} - \texttt{src1}(I) ) \quad \texttt{if mask}(I) \ne0\f] where I is a multi-dimensional index of array elements. In case of multi-channel arrays, each channel is processed independently.

The first function in the list above can be replaced with matrix expressions:

dst = src1 - src2;
dst -= src1; // equivalent to subtract(dst, src1, dst);

The input arrays and the output array can all have the same or different depths. For example, you can subtract to 8-bit unsigned arrays and store the difference in a 16-bit signed array. Depth of the output array is determined by dtype parameter. In the second and third cases above, as well as in the first case, when src1.depth() == src2.depth(), dtype can be set to the default -1. In this case the output array will have the same depth as the input array, be it src1, src2 or both. Note: Saturation is not applied when the output array has the depth CV_32S. You may even get result of an incorrect sign in the case of overflow. Note: (Python) Be careful to difference behaviour between src1/src2 are single number and they are tuple/array. subtract(src,X) means subtract(src,(X,X,X,X)). subtract(src,(X,)) means subtract(src,(X,0,0,0)). @sa add, addWeighted, scaleAdd, Mat::convertTo

Python prototype (for reference only):

subtract(src1, src2[, dst[, mask[, dtype]]]) -> dst
@spec sumElems(Evision.Mat.maybe_mat_in()) ::
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()}
  | {:error, String.t()}

Calculates the sum of array elements.

Positional Arguments
  • src: Evision.Mat.t().

    input array that must have from 1 to 4 channels.

Return
  • retval: Scalar

The function cv::sum calculates and returns the sum of array elements, independently for each channel. @sa countNonZero, mean, meanStdDev, norm, minMaxLoc, reduce

Python prototype (for reference only):

sumElems(src) -> retval
Link to this function

svBackSubst(w, u, vt, rhs)

View Source

SVBackSubst

Positional Arguments
  • w: Evision.Mat.t()
  • u: Evision.Mat.t()
  • vt: Evision.Mat.t()
  • rhs: Evision.Mat.t()
Return
  • dst: Evision.Mat.t().

wrap SVD::backSubst

Python prototype (for reference only):

SVBackSubst(w, u, vt, rhs[, dst]) -> dst
Link to this function

svBackSubst(w, u, vt, rhs, opts)

View Source

SVBackSubst

Positional Arguments
  • w: Evision.Mat.t()
  • u: Evision.Mat.t()
  • vt: Evision.Mat.t()
  • rhs: Evision.Mat.t()
Return
  • dst: Evision.Mat.t().

wrap SVD::backSubst

Python prototype (for reference only):

SVBackSubst(w, u, vt, rhs[, dst]) -> dst
@spec svdDecomp(Evision.Mat.maybe_mat_in()) ::
  {Evision.Mat.t(), Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

SVDecomp

Positional Arguments
  • src: Evision.Mat.t()
Keyword Arguments
  • flags: int.
Return
  • w: Evision.Mat.t().
  • u: Evision.Mat.t().
  • vt: Evision.Mat.t().

wrap SVD::compute

Python prototype (for reference only):

SVDecomp(src[, w[, u[, vt[, flags]]]]) -> w, u, vt
@spec svdDecomp(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  {Evision.Mat.t(), Evision.Mat.t(), Evision.Mat.t()} | {:error, String.t()}

SVDecomp

Positional Arguments
  • src: Evision.Mat.t()
Keyword Arguments
  • flags: int.
Return
  • w: Evision.Mat.t().
  • u: Evision.Mat.t().
  • vt: Evision.Mat.t().

wrap SVD::compute

Python prototype (for reference only):

SVDecomp(src[, w[, u[, vt[, flags]]]]) -> w, u, vt
Link to this function

textureFlattening(src, mask)

View Source
@spec textureFlattening(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

By retaining only the gradients at edge locations, before integrating with the Poisson solver, one washes out the texture of the selected region, giving its contents a flat aspect. Here Canny Edge %Detector is used.

Positional Arguments
  • src: Evision.Mat.t().

    Input 8-bit 3-channel image.

  • mask: Evision.Mat.t().

    Input 8-bit 1 or 3-channel image.

Keyword Arguments
  • low_threshold: float.

    %Range from 0 to 100.

  • high_threshold: float.

    Value > 100.

  • kernel_size: int.

    The size of the Sobel kernel to be used.

Return
  • dst: Evision.Mat.t().

    Output image with the same size and type as src.

Note: The algorithm assumes that the color of the source image is close to that of the destination. This assumption means that when the colors don't match, the source image color gets tinted toward the color of the destination image.

Python prototype (for reference only):

textureFlattening(src, mask[, dst[, low_threshold[, high_threshold[, kernel_size]]]]) -> dst
Link to this function

textureFlattening(src, mask, opts)

View Source
@spec textureFlattening(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

By retaining only the gradients at edge locations, before integrating with the Poisson solver, one washes out the texture of the selected region, giving its contents a flat aspect. Here Canny Edge %Detector is used.

Positional Arguments
  • src: Evision.Mat.t().

    Input 8-bit 3-channel image.

  • mask: Evision.Mat.t().

    Input 8-bit 1 or 3-channel image.

Keyword Arguments
  • low_threshold: float.

    %Range from 0 to 100.

  • high_threshold: float.

    Value > 100.

  • kernel_size: int.

    The size of the Sobel kernel to be used.

Return
  • dst: Evision.Mat.t().

    Output image with the same size and type as src.

Note: The algorithm assumes that the color of the source image is close to that of the destination. This assumption means that when the colors don't match, the source image color gets tinted toward the color of the destination image.

Python prototype (for reference only):

textureFlattening(src, mask[, dst[, low_threshold[, high_threshold[, kernel_size]]]]) -> dst
Link to this function

threshold(src, thresh, maxval, type)

View Source
@spec threshold(Evision.Mat.maybe_mat_in(), number(), number(), integer()) ::
  {number(), Evision.Mat.t()} | {:error, String.t()}

Applies a fixed-level threshold to each array element.

Positional Arguments
  • src: Evision.Mat.t().

    input array (multiple-channel, 8-bit or 32-bit floating point).

  • thresh: double.

    threshold value.

  • maxval: double.

    maximum value to use with the #THRESH_BINARY and #THRESH_BINARY_INV thresholding types.

  • type: int.

    thresholding type (see #ThresholdTypes).

Return
  • retval: double

  • dst: Evision.Mat.t().

    output array of the same size and type and the same number of channels as src.

The function applies fixed-level thresholding to a multiple-channel array. The function is typically used to get a bi-level (binary) image out of a grayscale image ( #compare could be also used for this purpose) or for removing a noise, that is, filtering out pixels with too small or too large values. There are several types of thresholding supported by the function. They are determined by type parameter. Also, the special values #THRESH_OTSU or #THRESH_TRIANGLE may be combined with one of the above values. In these cases, the function determines the optimal threshold value using the Otsu's or Triangle algorithm and uses it instead of the specified thresh. Note: Currently, the Otsu's and Triangle methods are implemented only for 8-bit single-channel images. @return the computed threshold value if Otsu's or Triangle methods used. @sa adaptiveThreshold, findContours, compare, min, max

Python prototype (for reference only):

threshold(src, thresh, maxval, type[, dst]) -> retval, dst
Link to this function

threshold(src, thresh, maxval, type, opts)

View Source
@spec threshold(
  Evision.Mat.maybe_mat_in(),
  number(),
  number(),
  integer(),
  [{atom(), term()}, ...] | nil
) :: {number(), Evision.Mat.t()} | {:error, String.t()}

Applies a fixed-level threshold to each array element.

Positional Arguments
  • src: Evision.Mat.t().

    input array (multiple-channel, 8-bit or 32-bit floating point).

  • thresh: double.

    threshold value.

  • maxval: double.

    maximum value to use with the #THRESH_BINARY and #THRESH_BINARY_INV thresholding types.

  • type: int.

    thresholding type (see #ThresholdTypes).

Return
  • retval: double

  • dst: Evision.Mat.t().

    output array of the same size and type and the same number of channels as src.

The function applies fixed-level thresholding to a multiple-channel array. The function is typically used to get a bi-level (binary) image out of a grayscale image ( #compare could be also used for this purpose) or for removing a noise, that is, filtering out pixels with too small or too large values. There are several types of thresholding supported by the function. They are determined by type parameter. Also, the special values #THRESH_OTSU or #THRESH_TRIANGLE may be combined with one of the above values. In these cases, the function determines the optimal threshold value using the Otsu's or Triangle algorithm and uses it instead of the specified thresh. Note: Currently, the Otsu's and Triangle methods are implemented only for 8-bit single-channel images. @return the computed threshold value if Otsu's or Triangle methods used. @sa adaptiveThreshold, findContours, compare, min, max

Python prototype (for reference only):

threshold(src, thresh, maxval, type[, dst]) -> retval, dst
@spec trace(Evision.Mat.maybe_mat_in()) ::
  {number()}
  | {number(), number()}
  | {number(), number(), number()}
  | {number(), number(), number(), number()}
  | {:error, String.t()}

Returns the trace of a matrix.

Positional Arguments
  • mtx: Evision.Mat.t().

    input matrix.

Return
  • retval: Scalar

The function cv::trace returns the sum of the diagonal elements of the matrix mtx . \f[\mathrm{tr} ( \texttt{mtx} ) = \sum _i \texttt{mtx} (i,i)\f]

Python prototype (for reference only):

trace(mtx) -> retval
@spec transform(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

Performs the matrix transformation of every array element.

Positional Arguments
  • src: Evision.Mat.t().

    input array that must have as many channels (1 to 4) as m.cols or m.cols-1.

  • m: Evision.Mat.t().

    transformation 2x2 or 2x3 floating-point matrix.

Return
  • dst: Evision.Mat.t().

    output array of the same size and depth as src; it has as many channels as m.rows.

The function cv::transform performs the matrix transformation of every element of the array src and stores the results in dst : \f[\texttt{dst} (I) = \texttt{m} \cdot \texttt{src} (I)\f] (when m.cols=src.channels() ), or \f[\texttt{dst} (I) = \texttt{m} \cdot [ \texttt{src} (I); 1]\f] (when m.cols=src.channels()+1 ) Every element of the N -channel array src is interpreted as N -element vector that is transformed using the M x N or M x (N+1) matrix m to M-element vector - the corresponding element of the output array dst . The function may be used for geometrical transformation of N -dimensional points, arbitrary linear color space transformation (such as various kinds of RGB to YUV transforms), shuffling the image channels, and so forth. @sa perspectiveTransform, getAffineTransform, estimateAffine2D, warpAffine, warpPerspective

Python prototype (for reference only):

transform(src, m[, dst]) -> dst
@spec transform(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Performs the matrix transformation of every array element.

Positional Arguments
  • src: Evision.Mat.t().

    input array that must have as many channels (1 to 4) as m.cols or m.cols-1.

  • m: Evision.Mat.t().

    transformation 2x2 or 2x3 floating-point matrix.

Return
  • dst: Evision.Mat.t().

    output array of the same size and depth as src; it has as many channels as m.rows.

The function cv::transform performs the matrix transformation of every element of the array src and stores the results in dst : \f[\texttt{dst} (I) = \texttt{m} \cdot \texttt{src} (I)\f] (when m.cols=src.channels() ), or \f[\texttt{dst} (I) = \texttt{m} \cdot [ \texttt{src} (I); 1]\f] (when m.cols=src.channels()+1 ) Every element of the N -channel array src is interpreted as N -element vector that is transformed using the M x N or M x (N+1) matrix m to M-element vector - the corresponding element of the output array dst . The function may be used for geometrical transformation of N -dimensional points, arbitrary linear color space transformation (such as various kinds of RGB to YUV transforms), shuffling the image channels, and so forth. @sa perspectiveTransform, getAffineTransform, estimateAffine2D, warpAffine, warpPerspective

Python prototype (for reference only):

transform(src, m[, dst]) -> dst
@spec transpose(Evision.Mat.maybe_mat_in()) :: Evision.Mat.t() | {:error, String.t()}

Transposes a matrix.

Positional Arguments
  • src: Evision.Mat.t().

    input array.

Return
  • dst: Evision.Mat.t().

    output array of the same type as src.

The function cv::transpose transposes the matrix src : \f[\texttt{dst} (i,j) = \texttt{src} (j,i)\f] Note: No complex conjugation is done in case of a complex matrix. It should be done separately if needed.

Python prototype (for reference only):

transpose(src[, dst]) -> dst
@spec transpose(Evision.Mat.maybe_mat_in(), [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Transposes a matrix.

Positional Arguments
  • src: Evision.Mat.t().

    input array.

Return
  • dst: Evision.Mat.t().

    output array of the same type as src.

The function cv::transpose transposes the matrix src : \f[\texttt{dst} (i,j) = \texttt{src} (j,i)\f] Note: No complex conjugation is done in case of a complex matrix. It should be done separately if needed.

Python prototype (for reference only):

transpose(src[, dst]) -> dst
@spec transposeND(Evision.Mat.maybe_mat_in(), [integer()]) ::
  Evision.Mat.t() | {:error, String.t()}

Transpose for n-dimensional matrices.

Positional Arguments
  • src: Evision.Mat.t().

    input array.

  • order: [int].

    a permutation of [0,1,..,N-1] where N is the number of axes of src. The i’th axis of dst will correspond to the axis numbered order[i] of the input.

Return
  • dst: Evision.Mat.t().

    output array of the same type as src.

Note: Input should be continuous single-channel matrix.

Python prototype (for reference only):

transposeND(src, order[, dst]) -> dst
Link to this function

transposeND(src, order, opts)

View Source
@spec transposeND(
  Evision.Mat.maybe_mat_in(),
  [integer()],
  [{atom(), term()}, ...] | nil
) ::
  Evision.Mat.t() | {:error, String.t()}

Transpose for n-dimensional matrices.

Positional Arguments
  • src: Evision.Mat.t().

    input array.

  • order: [int].

    a permutation of [0,1,..,N-1] where N is the number of axes of src. The i’th axis of dst will correspond to the axis numbered order[i] of the input.

Return
  • dst: Evision.Mat.t().

    output array of the same type as src.

Note: Input should be continuous single-channel matrix.

Python prototype (for reference only):

transposeND(src, order[, dst]) -> dst
Link to this function

triangulatePoints(projMatr1, projMatr2, projPoints1, projPoints2)

View Source

This function reconstructs 3-dimensional points (in homogeneous coordinates) by using their observations with a stereo camera.

Positional Arguments
  • projMatr1: Evision.Mat.t().

    3x4 projection matrix of the first camera, i.e. this matrix projects 3D points given in the world's coordinate system into the first image.

  • projMatr2: Evision.Mat.t().

    3x4 projection matrix of the second camera, i.e. this matrix projects 3D points given in the world's coordinate system into the second image.

  • projPoints1: Evision.Mat.t().

    2xN array of feature points in the first image. In the case of the c++ version, it can be also a vector of feature points or two-channel matrix of size 1xN or Nx1.

  • projPoints2: Evision.Mat.t().

    2xN array of corresponding points in the second image. In the case of the c++ version, it can be also a vector of feature points or two-channel matrix of size 1xN or Nx1.

Return
  • points4D: Evision.Mat.t().

    4xN array of reconstructed points in homogeneous coordinates. These points are returned in the world's coordinate system.

Note: Keep in mind that all input data should be of float type in order for this function to work. Note: If the projection matrices from @ref stereoRectify are used, then the returned points are represented in the first camera's rectified coordinate system. @sa reprojectImageTo3D

Python prototype (for reference only):

triangulatePoints(projMatr1, projMatr2, projPoints1, projPoints2[, points4D]) -> points4D
Link to this function

triangulatePoints(projMatr1, projMatr2, projPoints1, projPoints2, opts)

View Source
@spec triangulatePoints(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

This function reconstructs 3-dimensional points (in homogeneous coordinates) by using their observations with a stereo camera.

Positional Arguments
  • projMatr1: Evision.Mat.t().

    3x4 projection matrix of the first camera, i.e. this matrix projects 3D points given in the world's coordinate system into the first image.

  • projMatr2: Evision.Mat.t().

    3x4 projection matrix of the second camera, i.e. this matrix projects 3D points given in the world's coordinate system into the second image.

  • projPoints1: Evision.Mat.t().

    2xN array of feature points in the first image. In the case of the c++ version, it can be also a vector of feature points or two-channel matrix of size 1xN or Nx1.

  • projPoints2: Evision.Mat.t().

    2xN array of corresponding points in the second image. In the case of the c++ version, it can be also a vector of feature points or two-channel matrix of size 1xN or Nx1.

Return
  • points4D: Evision.Mat.t().

    4xN array of reconstructed points in homogeneous coordinates. These points are returned in the world's coordinate system.

Note: Keep in mind that all input data should be of float type in order for this function to work. Note: If the projection matrices from @ref stereoRectify are used, then the returned points are represented in the first camera's rectified coordinate system. @sa reprojectImageTo3D

Python prototype (for reference only):

triangulatePoints(projMatr1, projMatr2, projPoints1, projPoints2[, points4D]) -> points4D
Link to this function

undistort(src, cameraMatrix, distCoeffs)

View Source

Transforms an image to compensate for lens distortion.

Positional Arguments
  • src: Evision.Mat.t().

    Input (distorted) image.

  • cameraMatrix: Evision.Mat.t().

    Input camera matrix \f$A = \vecthreethree{f_x}{0}{c_x}{0}{f_y}{c_y}{0}{0}{1}\f$ .

  • distCoeffs: Evision.Mat.t().

    Input vector of distortion coefficients \f$(k_1, k_2, p_1, p_2[, k_3[, k_4, k_5, k_6[, s_1, s_2, s_3, s_4[, \tau_x, \tau_y]]]])\f$ of 4, 5, 8, 12 or 14 elements. If the vector is NULL/empty, the zero distortion coefficients are assumed.

Keyword Arguments
  • newCameraMatrix: Evision.Mat.t().

    Camera matrix of the distorted image. By default, it is the same as cameraMatrix but you may additionally scale and shift the result by using a different matrix.

Return
  • dst: Evision.Mat.t().

    Output (corrected) image that has the same size and type as src .

The function transforms an image to compensate radial and tangential lens distortion. The function is simply a combination of #initUndistortRectifyMap (with unity R ) and #remap (with bilinear interpolation). See the former function for details of the transformation being performed. Those pixels in the destination image, for which there is no correspondent pixels in the source image, are filled with zeros (black color). A particular subset of the source image that will be visible in the corrected image can be regulated by newCameraMatrix. You can use #getOptimalNewCameraMatrix to compute the appropriate newCameraMatrix depending on your requirements. The camera matrix and the distortion parameters can be determined using #calibrateCamera. If the resolution of images is different from the resolution used at the calibration stage, \f$f_x, f_y, c_x\f$ and \f$c_y\f$ need to be scaled accordingly, while the distortion coefficients remain the same.

Python prototype (for reference only):

undistort(src, cameraMatrix, distCoeffs[, dst[, newCameraMatrix]]) -> dst
Link to this function

undistort(src, cameraMatrix, distCoeffs, opts)

View Source
@spec undistort(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Transforms an image to compensate for lens distortion.

Positional Arguments
  • src: Evision.Mat.t().

    Input (distorted) image.

  • cameraMatrix: Evision.Mat.t().

    Input camera matrix \f$A = \vecthreethree{f_x}{0}{c_x}{0}{f_y}{c_y}{0}{0}{1}\f$ .

  • distCoeffs: Evision.Mat.t().

    Input vector of distortion coefficients \f$(k_1, k_2, p_1, p_2[, k_3[, k_4, k_5, k_6[, s_1, s_2, s_3, s_4[, \tau_x, \tau_y]]]])\f$ of 4, 5, 8, 12 or 14 elements. If the vector is NULL/empty, the zero distortion coefficients are assumed.

Keyword Arguments
  • newCameraMatrix: Evision.Mat.t().

    Camera matrix of the distorted image. By default, it is the same as cameraMatrix but you may additionally scale and shift the result by using a different matrix.

Return
  • dst: Evision.Mat.t().

    Output (corrected) image that has the same size and type as src .

The function transforms an image to compensate radial and tangential lens distortion. The function is simply a combination of #initUndistortRectifyMap (with unity R ) and #remap (with bilinear interpolation). See the former function for details of the transformation being performed. Those pixels in the destination image, for which there is no correspondent pixels in the source image, are filled with zeros (black color). A particular subset of the source image that will be visible in the corrected image can be regulated by newCameraMatrix. You can use #getOptimalNewCameraMatrix to compute the appropriate newCameraMatrix depending on your requirements. The camera matrix and the distortion parameters can be determined using #calibrateCamera. If the resolution of images is different from the resolution used at the calibration stage, \f$f_x, f_y, c_x\f$ and \f$c_y\f$ need to be scaled accordingly, while the distortion coefficients remain the same.

Python prototype (for reference only):

undistort(src, cameraMatrix, distCoeffs[, dst[, newCameraMatrix]]) -> dst
Link to this function

undistortImagePoints(src, cameraMatrix, distCoeffs)

View Source
@spec undistortImagePoints(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in()
) :: Evision.Mat.t() | {:error, String.t()}

Compute undistorted image points position

Positional Arguments
  • src: Evision.Mat.t().

    Observed points position, 2xN/Nx2 1-channel or 1xN/Nx1 2-channel (CV_32FC2 or CV_64FC2) (or vector\<Point2f> ).

  • cameraMatrix: Evision.Mat.t().

    Camera matrix \f$\vecthreethree{f_x}{0}{c_x}{0}{f_y}{c_y}{0}{0}{1}\f$ .

  • distCoeffs: Evision.Mat.t().

    Distortion coefficients

Keyword Arguments
  • arg1: TermCriteria.
Return
  • dst: Evision.Mat.t().

    Output undistorted points position (1xN/Nx1 2-channel or vector\<Point2f> ).

Python prototype (for reference only):

undistortImagePoints(src, cameraMatrix, distCoeffs[, dst[, arg1]]) -> dst
Link to this function

undistortImagePoints(src, cameraMatrix, distCoeffs, opts)

View Source
@spec undistortImagePoints(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Compute undistorted image points position

Positional Arguments
  • src: Evision.Mat.t().

    Observed points position, 2xN/Nx2 1-channel or 1xN/Nx1 2-channel (CV_32FC2 or CV_64FC2) (or vector\<Point2f> ).

  • cameraMatrix: Evision.Mat.t().

    Camera matrix \f$\vecthreethree{f_x}{0}{c_x}{0}{f_y}{c_y}{0}{0}{1}\f$ .

  • distCoeffs: Evision.Mat.t().

    Distortion coefficients

Keyword Arguments
  • arg1: TermCriteria.
Return
  • dst: Evision.Mat.t().

    Output undistorted points position (1xN/Nx1 2-channel or vector\<Point2f> ).

Python prototype (for reference only):

undistortImagePoints(src, cameraMatrix, distCoeffs[, dst[, arg1]]) -> dst
Link to this function

undistortPoints(src, cameraMatrix, distCoeffs)

View Source

Computes the ideal point coordinates from the observed point coordinates.

Positional Arguments
  • src: Evision.Mat.t().

    Observed point coordinates, 2xN/Nx2 1-channel or 1xN/Nx1 2-channel (CV_32FC2 or CV_64FC2) (or vector\<Point2f> ).

  • cameraMatrix: Evision.Mat.t().

    Camera matrix \f$\vecthreethree{f_x}{0}{c_x}{0}{f_y}{c_y}{0}{0}{1}\f$ .

  • distCoeffs: Evision.Mat.t().

    Input vector of distortion coefficients \f$(k_1, k_2, p_1, p_2[, k_3[, k_4, k_5, k_6[, s_1, s_2, s_3, s_4[, \tau_x, \tau_y]]]])\f$ of 4, 5, 8, 12 or 14 elements. If the vector is NULL/empty, the zero distortion coefficients are assumed.

Keyword Arguments
  • r: Evision.Mat.t().

    Rectification transformation in the object space (3x3 matrix). R1 or R2 computed by #stereoRectify can be passed here. If the matrix is empty, the identity transformation is used.

  • p: Evision.Mat.t().

    New camera matrix (3x3) or new projection matrix (3x4) \f$\begin{bmatrix} {f'}_x & 0 & {c'}_x & t_x \\ 0 & {f'}_y & {c'}_y & t_y \\ 0 & 0 & 1 & t_z \end{bmatrix}\f$. P1 or P2 computed by #stereoRectify can be passed here. If the matrix is empty, the identity new camera matrix is used.

Return
  • dst: Evision.Mat.t().

    Output ideal point coordinates (1xN/Nx1 2-channel or vector\<Point2f> ) after undistortion and reverse perspective transformation. If matrix P is identity or omitted, dst will contain normalized point coordinates.

The function is similar to #undistort and #initUndistortRectifyMap but it operates on a sparse set of points instead of a raster image. Also the function performs a reverse transformation to #projectPoints. In case of a 3D object, it does not reconstruct its 3D coordinates, but for a planar object, it does, up to a translation vector, if the proper R is specified. For each observed point coordinate \f$(u, v)\f$ the function computes: \f[ \begin{array}{l} x^{"} \leftarrow (u - c_x)/f_x \\ y^{"} \leftarrow (v - c_y)/f_y \\ (x',y') = undistort(x^{"},y^{"}, \texttt{distCoeffs}) \\ {[X\,Y\,W]} ^T \leftarrow R*[x' \, y' \, 1]^T \\ x \leftarrow X/W \\ y \leftarrow Y/W \\ \text{only performed if P is specified:} \\ u' \leftarrow x {f'}_x + {c'}_x \\ v' \leftarrow y {f'}_y + {c'}_y \end{array} \f] where undistort is an approximate iterative algorithm that estimates the normalized original point coordinates out of the normalized distorted point coordinates ("normalized" means that the coordinates do not depend on the camera matrix). The function can be used for both a stereo camera head or a monocular camera (when R is empty).

Python prototype (for reference only):

undistortPoints(src, cameraMatrix, distCoeffs[, dst[, R[, P]]]) -> dst
Link to this function

undistortPoints(src, cameraMatrix, distCoeffs, opts)

View Source
@spec undistortPoints(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Computes the ideal point coordinates from the observed point coordinates.

Positional Arguments
  • src: Evision.Mat.t().

    Observed point coordinates, 2xN/Nx2 1-channel or 1xN/Nx1 2-channel (CV_32FC2 or CV_64FC2) (or vector\<Point2f> ).

  • cameraMatrix: Evision.Mat.t().

    Camera matrix \f$\vecthreethree{f_x}{0}{c_x}{0}{f_y}{c_y}{0}{0}{1}\f$ .

  • distCoeffs: Evision.Mat.t().

    Input vector of distortion coefficients \f$(k_1, k_2, p_1, p_2[, k_3[, k_4, k_5, k_6[, s_1, s_2, s_3, s_4[, \tau_x, \tau_y]]]])\f$ of 4, 5, 8, 12 or 14 elements. If the vector is NULL/empty, the zero distortion coefficients are assumed.

Keyword Arguments
  • r: Evision.Mat.t().

    Rectification transformation in the object space (3x3 matrix). R1 or R2 computed by #stereoRectify can be passed here. If the matrix is empty, the identity transformation is used.

  • p: Evision.Mat.t().

    New camera matrix (3x3) or new projection matrix (3x4) \f$\begin{bmatrix} {f'}_x & 0 & {c'}_x & t_x \\ 0 & {f'}_y & {c'}_y & t_y \\ 0 & 0 & 1 & t_z \end{bmatrix}\f$. P1 or P2 computed by #stereoRectify can be passed here. If the matrix is empty, the identity new camera matrix is used.

Return
  • dst: Evision.Mat.t().

    Output ideal point coordinates (1xN/Nx1 2-channel or vector\<Point2f> ) after undistortion and reverse perspective transformation. If matrix P is identity or omitted, dst will contain normalized point coordinates.

The function is similar to #undistort and #initUndistortRectifyMap but it operates on a sparse set of points instead of a raster image. Also the function performs a reverse transformation to #projectPoints. In case of a 3D object, it does not reconstruct its 3D coordinates, but for a planar object, it does, up to a translation vector, if the proper R is specified. For each observed point coordinate \f$(u, v)\f$ the function computes: \f[ \begin{array}{l} x^{"} \leftarrow (u - c_x)/f_x \\ y^{"} \leftarrow (v - c_y)/f_y \\ (x',y') = undistort(x^{"},y^{"}, \texttt{distCoeffs}) \\ {[X\,Y\,W]} ^T \leftarrow R*[x' \, y' \, 1]^T \\ x \leftarrow X/W \\ y \leftarrow Y/W \\ \text{only performed if P is specified:} \\ u' \leftarrow x {f'}_x + {c'}_x \\ v' \leftarrow y {f'}_y + {c'}_y \end{array} \f] where undistort is an approximate iterative algorithm that estimates the normalized original point coordinates out of the normalized distorted point coordinates ("normalized" means that the coordinates do not depend on the camera matrix). The function can be used for both a stereo camera head or a monocular camera (when R is empty).

Python prototype (for reference only):

undistortPoints(src, cameraMatrix, distCoeffs[, dst[, R[, P]]]) -> dst
Link to this function

undistortPointsIter(src, cameraMatrix, distCoeffs, r, p, criteria)

View Source

undistortPointsIter

Positional Arguments
  • src: Evision.Mat.t()
  • cameraMatrix: Evision.Mat.t()
  • distCoeffs: Evision.Mat.t()
  • r: Evision.Mat.t()
  • p: Evision.Mat.t()
  • criteria: TermCriteria
Return
  • dst: Evision.Mat.t().

Has overloading in C++

Note: Default version of #undistortPoints does 5 iterations to compute undistorted points.

Python prototype (for reference only):

undistortPointsIter(src, cameraMatrix, distCoeffs, R, P, criteria[, dst]) -> dst
Link to this function

undistortPointsIter(src, cameraMatrix, distCoeffs, r, p, criteria, opts)

View Source

undistortPointsIter

Positional Arguments
  • src: Evision.Mat.t()
  • cameraMatrix: Evision.Mat.t()
  • distCoeffs: Evision.Mat.t()
  • r: Evision.Mat.t()
  • p: Evision.Mat.t()
  • criteria: TermCriteria
Return
  • dst: Evision.Mat.t().

Has overloading in C++

Note: Default version of #undistortPoints does 5 iterations to compute undistorted points.

Python prototype (for reference only):

undistortPointsIter(src, cameraMatrix, distCoeffs, R, P, criteria[, dst]) -> dst
@spec useOpenVX() :: boolean() | {:error, String.t()}

useOpenVX

Return
  • retval: bool

Python prototype (for reference only):

useOpenVX() -> retval
@spec useOptimized() :: boolean() | {:error, String.t()}

Returns the status of optimized code usage.

Return
  • retval: bool

The function returns true if the optimized code is enabled. Otherwise, it returns false.

Python prototype (for reference only):

useOptimized() -> retval
Link to this function

validateDisparity(disparity, cost, minDisparity, numberOfDisparities)

View Source
@spec validateDisparity(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  integer(),
  integer()
) ::
  Evision.Mat.t() | {:error, String.t()}

validateDisparity

Positional Arguments
  • cost: Evision.Mat.t()
  • minDisparity: int
  • numberOfDisparities: int
Keyword Arguments
  • disp12MaxDisp: int.
Return
  • disparity: Evision.Mat.t()

Python prototype (for reference only):

validateDisparity(disparity, cost, minDisparity, numberOfDisparities[, disp12MaxDisp]) -> disparity
Link to this function

validateDisparity(disparity, cost, minDisparity, numberOfDisparities, opts)

View Source
@spec validateDisparity(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  integer(),
  integer(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

validateDisparity

Positional Arguments
  • cost: Evision.Mat.t()
  • minDisparity: int
  • numberOfDisparities: int
Keyword Arguments
  • disp12MaxDisp: int.
Return
  • disparity: Evision.Mat.t()

Python prototype (for reference only):

validateDisparity(disparity, cost, minDisparity, numberOfDisparities[, disp12MaxDisp]) -> disparity
@spec vconcat([Evision.Mat.maybe_mat_in()]) :: Evision.Mat.t() | {:error, String.t()}

vconcat

Positional Arguments
  • src: [Evision.Mat].

    input array or vector of matrices. all of the matrices must have the same number of cols and the same depth

Return
  • dst: Evision.Mat.t().

    output array. It has the same number of cols and depth as the src, and the sum of rows of the src. same depth.

Has overloading in C++

std::vector<cv::Mat> matrices = { cv::Mat(1, 4, CV_8UC1, cv::Scalar(1)),
cv::Mat(1, 4, CV_8UC1, cv::Scalar(2)),
cv::Mat(1, 4, CV_8UC1, cv::Scalar(3)),};
cv::Mat out;
cv::vconcat( matrices, out );
//out:
//[1,   1,   1,   1;
// 2,   2,   2,   2;
// 3,   3,   3,   3]

Python prototype (for reference only):

vconcat(src[, dst]) -> dst
@spec vconcat([Evision.Mat.maybe_mat_in()], [{atom(), term()}, ...] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

vconcat

Positional Arguments
  • src: [Evision.Mat].

    input array or vector of matrices. all of the matrices must have the same number of cols and the same depth

Return
  • dst: Evision.Mat.t().

    output array. It has the same number of cols and depth as the src, and the sum of rows of the src. same depth.

Has overloading in C++

std::vector<cv::Mat> matrices = { cv::Mat(1, 4, CV_8UC1, cv::Scalar(1)),
cv::Mat(1, 4, CV_8UC1, cv::Scalar(2)),
cv::Mat(1, 4, CV_8UC1, cv::Scalar(3)),};
cv::Mat out;
cv::vconcat( matrices, out );
//out:
//[1,   1,   1,   1;
// 2,   2,   2,   2;
// 3,   3,   3,   3]

Python prototype (for reference only):

vconcat(src[, dst]) -> dst
@spec waitKeyEx() :: integer() | {:error, String.t()}

Similar to #waitKey, but returns full key code.

Keyword Arguments
  • delay: int.
Return
  • retval: int

Note: Key code is implementation specific and depends on used backend: QT/GTK/Win32/etc

Python prototype (for reference only):

waitKeyEx([, delay]) -> retval
@spec waitKeyEx([{atom(), term()}, ...] | nil) :: integer() | {:error, String.t()}

Similar to #waitKey, but returns full key code.

Keyword Arguments
  • delay: int.
Return
  • retval: int

Note: Key code is implementation specific and depends on used backend: QT/GTK/Win32/etc

Python prototype (for reference only):

waitKeyEx([, delay]) -> retval
Link to this function

warpAffine(src, m, dsize)

View Source
@spec warpAffine(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  {number(), number()}
) ::
  Evision.Mat.t() | {:error, String.t()}

Applies an affine transformation to an image.

Positional Arguments
  • src: Evision.Mat.t().

    input image.

  • m: Evision.Mat.t().

    \f$2\times 3\f$ transformation matrix.

  • dsize: Size.

    size of the output image.

Keyword Arguments
  • flags: int.

    combination of interpolation methods (see #InterpolationFlags) and the optional flag #WARP_INVERSE_MAP that means that M is the inverse transformation ( \f$\texttt{dst}\rightarrow\texttt{src}\f$ ).

  • borderMode: int.

    pixel extrapolation method (see #BorderTypes); when borderMode=#BORDER_TRANSPARENT, it means that the pixels in the destination image corresponding to the "outliers" in the source image are not modified by the function.

  • borderValue: Scalar.

    value used in case of a constant border; by default, it is 0.

Return
  • dst: Evision.Mat.t().

    output image that has the size dsize and the same type as src .

The function warpAffine transforms the source image using the specified matrix: \f[\texttt{dst} (x,y) = \texttt{src} ( \texttt{M} _{11} x + \texttt{M} _{12} y + \texttt{M} _{13}, \texttt{M} _{21} x + \texttt{M} _{22} y + \texttt{M} _{23})\f] when the flag #WARP_INVERSE_MAP is set. Otherwise, the transformation is first inverted with #invertAffineTransform and then put in the formula above instead of M. The function cannot operate in-place.

@sa warpPerspective, resize, remap, getRectSubPix, transform

Python prototype (for reference only):

warpAffine(src, M, dsize[, dst[, flags[, borderMode[, borderValue]]]]) -> dst
Link to this function

warpAffine(src, m, dsize, opts)

View Source
@spec warpAffine(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Applies an affine transformation to an image.

Positional Arguments
  • src: Evision.Mat.t().

    input image.

  • m: Evision.Mat.t().

    \f$2\times 3\f$ transformation matrix.

  • dsize: Size.

    size of the output image.

Keyword Arguments
  • flags: int.

    combination of interpolation methods (see #InterpolationFlags) and the optional flag #WARP_INVERSE_MAP that means that M is the inverse transformation ( \f$\texttt{dst}\rightarrow\texttt{src}\f$ ).

  • borderMode: int.

    pixel extrapolation method (see #BorderTypes); when borderMode=#BORDER_TRANSPARENT, it means that the pixels in the destination image corresponding to the "outliers" in the source image are not modified by the function.

  • borderValue: Scalar.

    value used in case of a constant border; by default, it is 0.

Return
  • dst: Evision.Mat.t().

    output image that has the size dsize and the same type as src .

The function warpAffine transforms the source image using the specified matrix: \f[\texttt{dst} (x,y) = \texttt{src} ( \texttt{M} _{11} x + \texttt{M} _{12} y + \texttt{M} _{13}, \texttt{M} _{21} x + \texttt{M} _{22} y + \texttt{M} _{23})\f] when the flag #WARP_INVERSE_MAP is set. Otherwise, the transformation is first inverted with #invertAffineTransform and then put in the formula above instead of M. The function cannot operate in-place.

@sa warpPerspective, resize, remap, getRectSubPix, transform

Python prototype (for reference only):

warpAffine(src, M, dsize[, dst[, flags[, borderMode[, borderValue]]]]) -> dst
Link to this function

warpPerspective(src, m, dsize)

View Source
@spec warpPerspective(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  {number(), number()}
) ::
  Evision.Mat.t() | {:error, String.t()}

Applies a perspective transformation to an image.

Positional Arguments
  • src: Evision.Mat.t().

    input image.

  • m: Evision.Mat.t().

    \f$3\times 3\f$ transformation matrix.

  • dsize: Size.

    size of the output image.

Keyword Arguments
  • flags: int.

    combination of interpolation methods (#INTER_LINEAR or #INTER_NEAREST) and the optional flag #WARP_INVERSE_MAP, that sets M as the inverse transformation ( \f$\texttt{dst}\rightarrow\texttt{src}\f$ ).

  • borderMode: int.

    pixel extrapolation method (#BORDER_CONSTANT or #BORDER_REPLICATE).

  • borderValue: Scalar.

    value used in case of a constant border; by default, it equals 0.

Return
  • dst: Evision.Mat.t().

    output image that has the size dsize and the same type as src .

The function warpPerspective transforms the source image using the specified matrix: \f[\texttt{dst} (x,y) = \texttt{src} \left ( \frac{M_{11} x + M_{12} y + M_{13}}{M_{31} x + M_{32} y + M_{33}} , \frac{M_{21} x + M_{22} y + M_{23}}{M_{31} x + M_{32} y + M_{33}} \right )\f] when the flag #WARP_INVERSE_MAP is set. Otherwise, the transformation is first inverted with invert and then put in the formula above instead of M. The function cannot operate in-place.

@sa warpAffine, resize, remap, getRectSubPix, perspectiveTransform

Python prototype (for reference only):

warpPerspective(src, M, dsize[, dst[, flags[, borderMode[, borderValue]]]]) -> dst
Link to this function

warpPerspective(src, m, dsize, opts)

View Source
@spec warpPerspective(
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

Applies a perspective transformation to an image.

Positional Arguments
  • src: Evision.Mat.t().

    input image.

  • m: Evision.Mat.t().

    \f$3\times 3\f$ transformation matrix.

  • dsize: Size.

    size of the output image.

Keyword Arguments
  • flags: int.

    combination of interpolation methods (#INTER_LINEAR or #INTER_NEAREST) and the optional flag #WARP_INVERSE_MAP, that sets M as the inverse transformation ( \f$\texttt{dst}\rightarrow\texttt{src}\f$ ).

  • borderMode: int.

    pixel extrapolation method (#BORDER_CONSTANT or #BORDER_REPLICATE).

  • borderValue: Scalar.

    value used in case of a constant border; by default, it equals 0.

Return
  • dst: Evision.Mat.t().

    output image that has the size dsize and the same type as src .

The function warpPerspective transforms the source image using the specified matrix: \f[\texttt{dst} (x,y) = \texttt{src} \left ( \frac{M_{11} x + M_{12} y + M_{13}}{M_{31} x + M_{32} y + M_{33}} , \frac{M_{21} x + M_{22} y + M_{23}}{M_{31} x + M_{32} y + M_{33}} \right )\f] when the flag #WARP_INVERSE_MAP is set. Otherwise, the transformation is first inverted with invert and then put in the formula above instead of M. The function cannot operate in-place.

@sa warpAffine, resize, remap, getRectSubPix, perspectiveTransform

Python prototype (for reference only):

warpPerspective(src, M, dsize[, dst[, flags[, borderMode[, borderValue]]]]) -> dst
Link to this function

warpPolar(src, dsize, center, maxRadius, flags)

View Source
@spec warpPolar(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  {number(), number()},
  number(),
  integer()
) :: Evision.Mat.t() | {:error, String.t()}

warpPolar

Positional Arguments
  • src: Evision.Mat.t().

    Source image.

  • dsize: Size.

    The destination image size (see description for valid options).

  • center: Point2f.

    The transformation center.

  • maxRadius: double.

    The radius of the bounding circle to transform. It determines the inverse magnitude scale parameter too.

  • flags: int.

    A combination of interpolation methods, #InterpolationFlags + #WarpPolarMode.

    • Add #WARP_POLAR_LINEAR to select linear polar mapping (default)
    • Add #WARP_POLAR_LOG to select semilog polar mapping
    • Add #WARP_INVERSE_MAP for reverse mapping.
Return
  • dst: Evision.Mat.t().

    Destination image. It will have same type as src.

\brief Remaps an image to polar or semilog-polar coordinates space @anchor polar_remaps_reference_image Polar remaps reference Transform the source image using the following transformation: \f[ dst(\rho , \phi ) = src(x,y) \f] where \f[ \begin{array}{l} \vec{I} = (x - center.x, \;y - center.y) \\ \phi = Kangle \cdot \texttt{angle} (\vec{I}) \\ \rho = \left\{\begin{matrix} Klin \cdot \texttt{magnitude} (\vec{I}) & default \\ Klog \cdot log_e(\texttt{magnitude} (\vec{I})) & if \; semilog \\ \end{matrix}\right. \end{array} \f] and \f[ \begin{array}{l} Kangle = dsize.height / 2\Pi \\ Klin = dsize.width / maxRadius \\ Klog = dsize.width / log_e(maxRadius) \\ \end{array} \f] \par Linear vs semilog mapping Polar mapping can be linear or semi-log. Add one of #WarpPolarMode to flags to specify the polar mapping mode. Linear is the default mode. The semilog mapping emulates the human "foveal" vision that permit very high acuity on the line of sight (central vision) in contrast to peripheral vision where acuity is minor. \par Option on dsize:

  • if both values in dsize <=0 (default), the destination image will have (almost) same area of source bounding circle: \f[\begin{array}{l} dsize.area \leftarrow (maxRadius^2 \cdot \Pi) \\ dsize.width = \texttt{cvRound}(maxRadius) \\ dsize.height = \texttt{cvRound}(maxRadius \cdot \Pi) \\ \end{array}\f]

  • if only dsize.height <= 0, the destination image area will be proportional to the bounding circle area but scaled by Kx * Kx: \f[\begin{array}{l} dsize.height = \texttt{cvRound}(dsize.width \cdot \Pi) \\ \end{array} \f]

  • if both values in dsize > 0, the destination image will have the given size therefore the area of the bounding circle will be scaled to dsize.

\par Reverse mapping You can get reverse mapping adding #WARP_INVERSE_MAP to flags \snippet polar_transforms.cpp InverseMap In addiction, to calculate the original coordinate from a polar mapped coordinate \f$(rho, phi)->(x, y)\f$: \snippet polar_transforms.cpp InverseCoordinate Note:

  • The function can not operate in-place.
  • To calculate magnitude and angle in degrees #cartToPolar is used internally thus angles are measured from 0 to 360 with accuracy about 0.3 degrees.
  • This function uses #remap. Due to current implementation limitations the size of an input and output images should be less than 32767x32767.

@sa cv::remap

Python prototype (for reference only):

warpPolar(src, dsize, center, maxRadius, flags[, dst]) -> dst
Link to this function

warpPolar(src, dsize, center, maxRadius, flags, opts)

View Source
@spec warpPolar(
  Evision.Mat.maybe_mat_in(),
  {number(), number()},
  {number(), number()},
  number(),
  integer(),
  [{atom(), term()}, ...] | nil
) :: Evision.Mat.t() | {:error, String.t()}

warpPolar

Positional Arguments
  • src: Evision.Mat.t().

    Source image.

  • dsize: Size.

    The destination image size (see description for valid options).

  • center: Point2f.

    The transformation center.

  • maxRadius: double.

    The radius of the bounding circle to transform. It determines the inverse magnitude scale parameter too.

  • flags: int.

    A combination of interpolation methods, #InterpolationFlags + #WarpPolarMode.

    • Add #WARP_POLAR_LINEAR to select linear polar mapping (default)
    • Add #WARP_POLAR_LOG to select semilog polar mapping
    • Add #WARP_INVERSE_MAP for reverse mapping.
Return
  • dst: Evision.Mat.t().

    Destination image. It will have same type as src.

\brief Remaps an image to polar or semilog-polar coordinates space @anchor polar_remaps_reference_image Polar remaps reference Transform the source image using the following transformation: \f[ dst(\rho , \phi ) = src(x,y) \f] where \f[ \begin{array}{l} \vec{I} = (x - center.x, \;y - center.y) \\ \phi = Kangle \cdot \texttt{angle} (\vec{I}) \\ \rho = \left\{\begin{matrix} Klin \cdot \texttt{magnitude} (\vec{I}) & default \\ Klog \cdot log_e(\texttt{magnitude} (\vec{I})) & if \; semilog \\ \end{matrix}\right. \end{array} \f] and \f[ \begin{array}{l} Kangle = dsize.height / 2\Pi \\ Klin = dsize.width / maxRadius \\ Klog = dsize.width / log_e(maxRadius) \\ \end{array} \f] \par Linear vs semilog mapping Polar mapping can be linear or semi-log. Add one of #WarpPolarMode to flags to specify the polar mapping mode. Linear is the default mode. The semilog mapping emulates the human "foveal" vision that permit very high acuity on the line of sight (central vision) in contrast to peripheral vision where acuity is minor. \par Option on dsize:

  • if both values in dsize <=0 (default), the destination image will have (almost) same area of source bounding circle: \f[\begin{array}{l} dsize.area \leftarrow (maxRadius^2 \cdot \Pi) \\ dsize.width = \texttt{cvRound}(maxRadius) \\ dsize.height = \texttt{cvRound}(maxRadius \cdot \Pi) \\ \end{array}\f]

  • if only dsize.height <= 0, the destination image area will be proportional to the bounding circle area but scaled by Kx * Kx: \f[\begin{array}{l} dsize.height = \texttt{cvRound}(dsize.width \cdot \Pi) \\ \end{array} \f]

  • if both values in dsize > 0, the destination image will have the given size therefore the area of the bounding circle will be scaled to dsize.

\par Reverse mapping You can get reverse mapping adding #WARP_INVERSE_MAP to flags \snippet polar_transforms.cpp InverseMap In addiction, to calculate the original coordinate from a polar mapped coordinate \f$(rho, phi)->(x, y)\f$: \snippet polar_transforms.cpp InverseCoordinate Note:

  • The function can not operate in-place.
  • To calculate magnitude and angle in degrees #cartToPolar is used internally thus angles are measured from 0 to 360 with accuracy about 0.3 degrees.
  • This function uses #remap. Due to current implementation limitations the size of an input and output images should be less than 32767x32767.

@sa cv::remap

Python prototype (for reference only):

warpPolar(src, dsize, center, maxRadius, flags[, dst]) -> dst
Link to this function

watershed(image, markers)

View Source
@spec watershed(Evision.Mat.maybe_mat_in(), Evision.Mat.maybe_mat_in()) ::
  Evision.Mat.t() | {:error, String.t()}

Performs a marker-based image segmentation using the watershed algorithm.

Positional Arguments
  • image: Evision.Mat.t().

    Input 8-bit 3-channel image.

Return
  • markers: Evision.Mat.t().

    Input/output 32-bit single-channel image (map) of markers. It should have the same size as image .

The function implements one of the variants of watershed, non-parametric marker-based segmentation algorithm, described in @cite Meyer92 . Before passing the image to the function, you have to roughly outline the desired regions in the image markers with positive (>0) indices. So, every region is represented as one or more connected components with the pixel values 1, 2, 3, and so on. Such markers can be retrieved from a binary mask using #findContours and #drawContours (see the watershed.cpp demo). The markers are "seeds" of the future image regions. All the other pixels in markers , whose relation to the outlined regions is not known and should be defined by the algorithm, should be set to 0's. In the function output, each pixel in markers is set to a value of the "seed" components or to -1 at boundaries between the regions. Note: Any two neighbor connected components are not necessarily separated by a watershed boundary (-1's pixels); for example, they can touch each other in the initial marker image passed to the function.

@sa findContours

Python prototype (for reference only):

watershed(image, markers) -> markers
Link to this function

writeOpticalFlow(path, flow)

View Source
@spec writeOpticalFlow(binary(), Evision.Mat.maybe_mat_in()) ::
  boolean() | {:error, String.t()}

Write a .flo to disk

Positional Arguments
  • path: String.

    Path to the file to be written

  • flow: Evision.Mat.t().

    Flow field to be stored

Return
  • retval: bool

The function stores a flow field in a file, returns true on success, false otherwise. The flow field must be a 2-channel, floating-point matrix (CV_32FC2). First channel corresponds to the flow in the horizontal direction (u), second - vertical (v).

Python prototype (for reference only):

writeOpticalFlow(path, flow) -> retval