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

Specify input image and extract image features

Positional Arguments

• self: Evision.Segmentation.IntelligentScissorsMB.t()

• image: Evision.Mat.

  input image. Type is #CV_8UC1 / #CV_8UC3

Return

• retval: Evision.Segmentation.IntelligentScissorsMB.t()

Python prototype (for reference only):

applyImage(image) -> retval

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

Specify custom features of input image

Positional Arguments

• self: Evision.Segmentation.IntelligentScissorsMB.t()

• non_edge: Evision.Mat.

  Specify cost of non-edge pixels. Type is CV_8UC1. Expected values are {0, 1}.

• gradient_direction: Evision.Mat.

  Specify gradient direction feature. Type is CV_32FC2. Values are expected to be normalized: x^2 + y^2 == 1

• gradient_magnitude: Evision.Mat.

  Specify cost of gradient magnitude function: Type is CV_32FC1. Values should be in range [0, 1].

Keyword Arguments

• image: Evision.Mat.

  Optional parameter. Must be specified if subset of features is specified (non-specified features are calculated internally)

Return

• retval: Evision.Segmentation.IntelligentScissorsMB.t()

Customized advanced variant of applyImage() call.

Python prototype (for reference only):

applyImageFeatures(non_edge, gradient_direction, gradient_magnitude[, image]) -> retval

@spec applyImageFeatures(
  t(),
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  Evision.Mat.maybe_mat_in(),
  [{:image, term()}] | nil
) :: t() | {:error, String.t()}

Specify custom features of input image

Positional Arguments

• self: Evision.Segmentation.IntelligentScissorsMB.t()

• non_edge: Evision.Mat.

  Specify cost of non-edge pixels. Type is CV_8UC1. Expected values are {0, 1}.

• gradient_direction: Evision.Mat.

  Specify gradient direction feature. Type is CV_32FC2. Values are expected to be normalized: x^2 + y^2 == 1

• gradient_magnitude: Evision.Mat.

  Specify cost of gradient magnitude function: Type is CV_32FC1. Values should be in range [0, 1].

Keyword Arguments

• image: Evision.Mat.

  Optional parameter. Must be specified if subset of features is specified (non-specified features are calculated internally)

Return

• retval: Evision.Segmentation.IntelligentScissorsMB.t()

Customized advanced variant of applyImage() call.

Python prototype (for reference only):

applyImageFeatures(non_edge, gradient_direction, gradient_magnitude[, image]) -> retval

@spec buildMap(
  t(),
  {number(), number()}
) :: t() | {:error, String.t()}

Prepares a map of optimal paths for the given source point on the image

Positional Arguments

• self: Evision.Segmentation.IntelligentScissorsMB.t()

• sourcePt: Point.

  The source point used to find the paths

Note: applyImage() / applyImageFeatures() must be called before this call

Python prototype (for reference only):

buildMap(sourcePt) -> None

@spec getContour(
  t(),
  {number(), number()}
) :: Evision.Mat.t() | {:error, String.t()}

Extracts optimal contour for the given target point on the image

Positional Arguments

• self: Evision.Segmentation.IntelligentScissorsMB.t()

• targetPt: Point.

  The target point

Keyword Arguments

• backward: bool.

  Flag to indicate reverse order of retrived pixels (use "true" value to fetch points from the target to the source point)

Return

• contour: Evision.Mat.t().

  The list of pixels which contains optimal path between the source and the target points of the image. Type is CV_32SC2 (compatible with std::vector<Point>)

Note: buildMap() must be called before this call

Python prototype (for reference only):

getContour(targetPt[, contour[, backward]]) -> contour

@spec getContour(t(), {number(), number()}, [{:backward, term()}] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Extracts optimal contour for the given target point on the image

Positional Arguments

• self: Evision.Segmentation.IntelligentScissorsMB.t()

• targetPt: Point.

  The target point

Keyword Arguments

• backward: bool.

  Flag to indicate reverse order of retrived pixels (use "true" value to fetch points from the target to the source point)

Return

• contour: Evision.Mat.t().

  The list of pixels which contains optimal path between the source and the target points of the image. Type is CV_32SC2 (compatible with std::vector<Point>)

Note: buildMap() must be called before this call

Python prototype (for reference only):

getContour(targetPt[, contour[, backward]]) -> contour

@spec intelligentScissorsMB() :: t() | {:error, String.t()}

IntelligentScissorsMB

Return

• self: Evision.Segmentation.IntelligentScissorsMB.t()

Python prototype (for reference only):

IntelligentScissorsMB() -> <segmentation_IntelligentScissorsMB object>

@spec setEdgeFeatureCannyParameters(t(), number(), number()) ::
  t() | {:error, String.t()}

Switch edge feature extractor to use Canny edge detector

Positional Arguments

• self: Evision.Segmentation.IntelligentScissorsMB.t()
• threshold1: double
• threshold2: double

Keyword Arguments

• apertureSize: integer().
• l2gradient: bool.

Return

• retval: Evision.Segmentation.IntelligentScissorsMB.t()

Note: "Laplacian Zero-Crossing" feature extractor is used by default (following to original article) @sa Canny

Python prototype (for reference only):

setEdgeFeatureCannyParameters(threshold1, threshold2[, apertureSize[, L2gradient]]) -> retval

@spec setEdgeFeatureCannyParameters(
  t(),
  number(),
  number(),
  [apertureSize: term(), l2gradient: term()] | nil
) :: t() | {:error, String.t()}

Switch edge feature extractor to use Canny edge detector

Positional Arguments

• self: Evision.Segmentation.IntelligentScissorsMB.t()
• threshold1: double
• threshold2: double

Keyword Arguments

• apertureSize: integer().
• l2gradient: bool.

Return

• retval: Evision.Segmentation.IntelligentScissorsMB.t()

Note: "Laplacian Zero-Crossing" feature extractor is used by default (following to original article) @sa Canny

Python prototype (for reference only):

setEdgeFeatureCannyParameters(threshold1, threshold2[, apertureSize[, L2gradient]]) -> retval

@spec setEdgeFeatureZeroCrossingParameters(Keyword.t()) ::
  any() | {:error, String.t()}

@spec setEdgeFeatureZeroCrossingParameters(t()) :: t() | {:error, String.t()}

Switch to "Laplacian Zero-Crossing" edge feature extractor and specify its parameters

Positional Arguments

• self: Evision.Segmentation.IntelligentScissorsMB.t()

Keyword Arguments

• gradient_magnitude_min_value: float.

  Minimal gradient magnitude value for edge pixels (default: 0, check is disabled)

Return

• retval: Evision.Segmentation.IntelligentScissorsMB.t()

This feature extractor is used by default according to article. Implementation has additional filtering for regions with low-amplitude noise. This filtering is enabled through parameter of minimal gradient amplitude (use some small value 4, 8, 16). Note: Current implementation of this feature extractor is based on processing of grayscale images (color image is converted to grayscale image first). Note: Canny edge detector is a bit slower, but provides better results (especially on color images): use setEdgeFeatureCannyParameters().

Python prototype (for reference only):

setEdgeFeatureZeroCrossingParameters([, gradient_magnitude_min_value]) -> retval

@spec setEdgeFeatureZeroCrossingParameters(
  t(),
  [{:gradient_magnitude_min_value, term()}] | nil
) ::
  t() | {:error, String.t()}

Switch to "Laplacian Zero-Crossing" edge feature extractor and specify its parameters

Positional Arguments

• self: Evision.Segmentation.IntelligentScissorsMB.t()

Keyword Arguments

• gradient_magnitude_min_value: float.

  Minimal gradient magnitude value for edge pixels (default: 0, check is disabled)

Return

• retval: Evision.Segmentation.IntelligentScissorsMB.t()

This feature extractor is used by default according to article. Implementation has additional filtering for regions with low-amplitude noise. This filtering is enabled through parameter of minimal gradient amplitude (use some small value 4, 8, 16). Note: Current implementation of this feature extractor is based on processing of grayscale images (color image is converted to grayscale image first). Note: Canny edge detector is a bit slower, but provides better results (especially on color images): use setEdgeFeatureCannyParameters().

Python prototype (for reference only):

setEdgeFeatureZeroCrossingParameters([, gradient_magnitude_min_value]) -> retval

@spec setGradientMagnitudeMaxLimit(Keyword.t()) :: any() | {:error, String.t()}

@spec setGradientMagnitudeMaxLimit(t()) :: t() | {:error, String.t()}

Specify gradient magnitude max value threshold

Positional Arguments

• self: Evision.Segmentation.IntelligentScissorsMB.t()

Keyword Arguments

• gradient_magnitude_threshold_max: float.

  Specify gradient magnitude max value threshold (default: 0, disabled)

Return

• retval: Evision.Segmentation.IntelligentScissorsMB.t()

Zero limit value is used to disable gradient magnitude thresholding (default behavior, as described in original article). Otherwize pixels with gradient magnitude >= threshold have zero cost. Note: Thresholding should be used for images with irregular regions (to avoid stuck on parameters from high-contract areas, like embedded logos).

Python prototype (for reference only):

setGradientMagnitudeMaxLimit([, gradient_magnitude_threshold_max]) -> retval

@spec setGradientMagnitudeMaxLimit(
  t(),
  [{:gradient_magnitude_threshold_max, term()}] | nil
) ::
  t() | {:error, String.t()}

Specify gradient magnitude max value threshold

Positional Arguments

• self: Evision.Segmentation.IntelligentScissorsMB.t()

Keyword Arguments

• gradient_magnitude_threshold_max: float.

  Specify gradient magnitude max value threshold (default: 0, disabled)

Return

• retval: Evision.Segmentation.IntelligentScissorsMB.t()

Zero limit value is used to disable gradient magnitude thresholding (default behavior, as described in original article). Otherwize pixels with gradient magnitude >= threshold have zero cost. Note: Thresholding should be used for images with irregular regions (to avoid stuck on parameters from high-contract areas, like embedded logos).

Python prototype (for reference only):

setGradientMagnitudeMaxLimit([, gradient_magnitude_threshold_max]) -> retval

@spec setWeights(t(), number(), number(), number()) :: t() | {:error, String.t()}

Specify weights of feature functions

Positional Arguments

• self: Evision.Segmentation.IntelligentScissorsMB.t()

• weight_non_edge: float.

  Specify cost of non-edge pixels (default: 0.43f)

• weight_gradient_direction: float.

  Specify cost of gradient direction function (default: 0.43f)

• weight_gradient_magnitude: float.

  Specify cost of gradient magnitude function (default: 0.14f)

Return

• retval: Evision.Segmentation.IntelligentScissorsMB.t()

Consider keeping weights normalized (sum of weights equals to 1.0) Discrete dynamic programming (DP) goal is minimization of costs between pixels.

Python prototype (for reference only):

setWeights(weight_non_edge, weight_gradient_direction, weight_gradient_magnitude) -> retval