View Source Evision.BOWImgDescriptorExtractor (Evision v0.1.21)

Link to this section Summary

Types

t()

Type that represents an Evision.BOWImgDescriptorExtractor struct.

Functions

Returns an image descriptor size if the vocabulary is set. Otherwise, it returns 0.

Returns an image descriptor type.

Returns the set vocabulary.

Sets a visual vocabulary.

Link to this section Types

@type t() :: %Evision.BOWImgDescriptorExtractor{ref: reference()}

Type that represents an Evision.BOWImgDescriptorExtractor struct.

  • ref. reference()

    The underlying erlang resource variable.

Link to this section Functions

Link to this function

bowImgDescriptorExtractor(dextractor, dmatcher)

View Source
@spec bowImgDescriptorExtractor(reference() | term(), Evision.DescriptorMatcher.t()) ::
  t() | {:error, String.t()}

The constructor.

Positional Arguments
  • dextractor: Ptr<DescriptorExtractor>.

    Descriptor extractor that is used to compute descriptors for an input image and its keypoints.

  • dmatcher: Evision.DescriptorMatcher.

    Descriptor matcher that is used to find the nearest word of the trained vocabulary for each keypoint descriptor of the image.

Return

Python prototype (for reference only):

BOWImgDescriptorExtractor(dextractor, dmatcher) -> <BOWImgDescriptorExtractor object>
Link to this function

compute(self, image, keypoints)

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

compute

Positional Arguments
  • self: Evision.BOWImgDescriptorExtractor.t()
  • image: Evision.Mat
  • keypoints: [Evision.KeyPoint]
Return
  • imgDescriptor: Evision.Mat.

    Computed output image descriptor.

Has overloading in C++

Python prototype (for reference only):

compute(image, keypoints[, imgDescriptor]) -> imgDescriptor
Link to this function

compute(self, image, keypoints, opts)

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

compute

Positional Arguments
  • self: Evision.BOWImgDescriptorExtractor.t()
  • image: Evision.Mat
  • keypoints: [Evision.KeyPoint]
Return
  • imgDescriptor: Evision.Mat.

    Computed output image descriptor.

Has overloading in C++

Python prototype (for reference only):

compute(image, keypoints[, imgDescriptor]) -> imgDescriptor
@spec descriptorSize(t()) :: integer() | {:error, String.t()}

Returns an image descriptor size if the vocabulary is set. Otherwise, it returns 0.

Positional Arguments
  • self: Evision.BOWImgDescriptorExtractor.t()
Return
  • retval: int

Python prototype (for reference only):

descriptorSize() -> retval
@spec descriptorType(t()) :: integer() | {:error, String.t()}

Returns an image descriptor type.

Positional Arguments
  • self: Evision.BOWImgDescriptorExtractor.t()
Return
  • retval: int

Python prototype (for reference only):

descriptorType() -> retval
@spec getVocabulary(t()) :: Evision.Mat.t() | {:error, String.t()}

Returns the set vocabulary.

Positional Arguments
  • self: Evision.BOWImgDescriptorExtractor.t()
Return

Python prototype (for reference only):

getVocabulary() -> retval
Link to this function

setVocabulary(self, vocabulary)

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

Sets a visual vocabulary.

Positional Arguments
  • self: Evision.BOWImgDescriptorExtractor.t()

  • vocabulary: Evision.Mat.

    Vocabulary (can be trained using the inheritor of BOWTrainer ). Each row of the vocabulary is a visual word (cluster center).

Python prototype (for reference only):

setVocabulary(vocabulary) -> None