Computes image gradients. More...

Functions
template<class D >
Image< D >	Canny (const Image< D > &suppressed_sobel, float lower_threshold, float higher_threshold)
	Returns Canny edge detection image. Lower and higher threshold as frequencies from 0.0 to 1.0. More...

template<class D >
Image< D >	Canny (const Image< D > &img, float lower_threshold, float higher_threshold, float sigma)
	Returns Canny edge detection image. Lower and higher threshold as frequencies from 0.0 to 1.0. More...

template<class D >
Image< D >	NonMaxSobelSuppression (const Image< D > &img)
	Returns the sobel magnitude image with non-maximum intensities suppressed. More...

template<class D >
Image< D >	NonMaxSobelSuppression (const Image< D > &magnitude, const Image< int > &direction)
	Returns the sobel magnitude image with non-maximum intensities suppressed. More...

template<class D >
Image< D >	Directional (const Image< D > &img, const Adjacency &adj)
	Returns a gradient composed by the sum of differences between a pixel and its adjacents. More...

template<class D >
void	Gabor (const Image< D > &img, float sigma, Image< D > magnitude=nullptr, Image< int > direction=nullptr)
	Returns the directional Gabor filter over image. More...

template<class D >
Image< D >	Morphological (const Image< D > &image)
	Computes the gradient image (dilation - erosion) using spherical adjacency relation of one pixel of radius. More...

template<class D >
Image< D >	Morphological (const Image< D > &image, const Adjacency &adjacency)
	Computes the gradient image. More...

template<class D >
Image< D >	CloseHoles (const Image< D > &canny)
	Returns Canny gradient with one pixel wide holes closed. More...

template<class D >
void	RemoveRedundantPoint (Image< D > &canny)
	Removes redundant points from components leaving the minimal width components. More...

template<class D >
void	RemoveBifurcation (Image< D > &canny)
	Removes points of bifurcation. It tries to keep straighter lines when possible. More...

Vector< size_t >	Extremities (const Image< int > &label)
	Gets one of the extremities of each segment. More...

template<class D >
void	RemoveSmallSegments (Image< D > &canny, size_t min_seg_size)
	Removes the edges that are smaller than the given size. More...

template<class D >
Image< D >	ScaleCanny (const Image< D > &img, size_t window_scale, float lower_threshold, float higher_threshold)
	Returns a scaled Canny edge detection image. Lower and higher threshold as frequencies from 0.0 to 1.0. More...

template<class D >
Image< D >	SubScaleCanny (const Image< D > &img, const Image< D > &base_grad, size_t window_scale, float lower_threshold)
	Returns a sub scaled Canny edge detection image. Lower threshold as a frequency from 0.0 to 1.0. More...

template<class D >
Image< D >	MultiScaleCanny (const Image< D > &img, float lower_threshold, float higher_threshold, float sigma, size_t scales)
	Returns multi scale Canny edge detection image. Lower and higher threshold as frequencies from 0.0 to 1.0. More...

template<class D >
Image< D >	MultiSubScaleCanny (const Image< D > &img, float lower_threshold, float higher_threshold, float sigma, size_t scales)
	Returns multi scale Canny edge detection image. Lower and higher threshold as frequencies from 0.0 to 1.0. More...

template<class D >
Image< D >	MultiScaleSizeCanny (const Image< D > &img, float lower_threshold, float higher_threshold, float sigma, size_t scales, float fraction)
	Returns multi scale Canny edge detection image. Lower and higher threshold as frequencies from 0.0 to 1.0. More...

template<class D >
Image< D >	DirectionalSobel (const Image< D > &img, size_t direction)
	Returns the directional Sobel filter over image. More...

template<class D >
void	Sobel (const Image< D > &img, Image< D > magnitude=nullptr, Image< int > direction=nullptr)
	Returns the directional Sobel filter over image. More...

Detailed Description

Computes image gradients.

Function Documentation

template<class D >

Image< D > Bial::Gradient::Canny	(	const Image< D > &	suppressed_sobel,
		float	lower_threshold,
		float	higher_threshold
	)

Returns Canny edge detection image. Lower and higher threshold as frequencies from 0.0 to 1.0.

Date: 2013/Nov/21

Parameters

suppressed_sobel	Suppressed gradient image given.
lower_threshold	Lower hysteresis threshold.
higher_threshold	Higher hysteresis threshold.

Returns: Canny edge detection image.

Warning: none.

template<class D >

Image< D > Bial::Gradient::Canny	(	const Image< D > &	img,
		float	lower_threshold,
		float	higher_threshold,
		float	sigma
	)

Returns Canny edge detection image. Lower and higher threshold as frequencies from 0.0 to 1.0.

Date: 2013/Nov/21

Parameters

img	Input image.
lower_threshold	Lower hysteresis threshold.
higher_threshold	Higher hysteresis threshold.
sigma	Standard deviation of Gaussian filter.

Returns: Canny edge detection image.

Warning: none.

template<class D >

Image< D > Bial::Gradient::CloseHoles ( const Image< D > & canny )

Returns Canny gradient with one pixel wide holes closed.

Date: 2015/Apr/10

Parameters

canny Input Canny gradient image.

Returns: none.

Warning: none.

template<class D >

Image< D > Bial::Gradient::Directional	(	const Image< D > &	img,
		const Adjacency &	adj
	)

Returns a gradient composed by the sum of differences between a pixel and its adjacents.

Date: 2013/Dec/19

Parameters

img	Input image.
adj	An adjacency relation.

Returns: Gradient composed by the sum of differences between a pixel and its adjacents.

Warning: none.

template<class D >

Image< D > Bial::Gradient::DirectionalSobel	(	const Image< D > &	img,
		size_t	direction
	)

Returns the directional Sobel filter over image.

Date: 2013/Nov/22

Parameters

img	Input image.
direction	One of the dimensions of the input image.

Returns: Directional Sobel filter over image.

Warning: none.

Vector< size_t > Bial::Gradient::Extremities ( const Image< int > & label )

Gets one of the extremities of each segment.

Date: 2015/Jul/08

Parameters

label An image containing labeled Canny gradient segments.

Returns: Vector with pixel indexs containing one extremity of each component.

Warning: Only implemented for 2D images.

template<class D >

void Bial::Gradient::Gabor	(	const Image< D > &	img,
		float	sigma,
		Image< D > *	magnitude = `nullptr`,
		Image< int > *	direction = `nullptr`
	)

Returns the directional Gabor filter over image.

Date: 2015/Jun/03

Parameters

img	Input image.
sigma	Standard deviation of the Gaussian.
magnitude	Gabor magnitude image.
direction	Output Gabor direction image.

Returns: Directional Gabor filter over image.

Warning: none.

template<class D >

Image< D > Bial::Gradient::Morphological ( const Image< D > & image )

Computes the gradient image (dilation - erosion) using spherical adjacency relation of one pixel of radius.

Date: 2012/Sep/21

Parameters

image Input image.

Returns: Gradient image.

Warning: none.

template<class D >

Image< D > Bial::Gradient::Morphological	(	const Image< D > &	image,
		const Adjacency &	adjacency
	)

Computes the gradient image.

Date: 2012/Sep/21

Parameters

image	Input image.
adjacency	Adjacency relation used by dilation and erosion opperations.

Returns: Gradient image.

Warning: none.

template<class D >

Image< D > Bial::Gradient::MultiScaleCanny	(	const Image< D > &	img,
		float	lower_threshold,
		float	higher_threshold,
		float	sigma,
		size_t	scales
	)

Returns multi scale Canny edge detection image. Lower and higher threshold as frequencies from 0.0 to 1.0.

Date: 2015/Jan/26

Parameters

img	Input image.
lower_threshold	Lower hysteresis threshold.
higher_threshold	Higher hysteresis threshold.
sigma	Standard deviation of Gaussian filter.
scales	Number of scales to run Canny.

Returns: Multi scale Canny edge detection image.

Warning: none.

template<class D >

Image< D > Bial::Gradient::MultiScaleSizeCanny	(	const Image< D > &	img,
		float	lower_threshold,
		float	higher_threshold,
		float	sigma,
		size_t	scales,
		float	fraction
	)

Returns multi scale Canny edge detection image. Lower and higher threshold as frequencies from 0.0 to 1.0.

Date: 2015/Jun/02

Parameters

img	Input image.
lower_threshold	Lower hysteresis threshold.
higher_threshold	Higher hysteresis threshold.
sigma	Standard deviation of Gaussian filter.
scales	Number of scales to run Canny.
fraction	Fraction of the maximum size edge that specifies the minimum edge size. (0.0, 1.0)

Returns: Multi scale Canny edge detection image.

Warning: none.

template<class D >

Image< D > Bial::Gradient::MultiSubScaleCanny	(	const Image< D > &	img,
		float	lower_threshold,
		float	higher_threshold,
		float	sigma,
		size_t	scales
	)

Returns multi scale Canny edge detection image. Lower and higher threshold as frequencies from 0.0 to 1.0.

Date: 2015/Jan/27

Parameters

img	Input image.
lower_threshold	Lower hysteresis threshold.
higher_threshold	Higher hysteresis threshold.
sigma	Standard deviation of Gaussian filter.
scales	Number of scales to run Canny.

Returns: Multi scale Canny edge detection image.

Warning: none.

template<class D >

Image< D > Bial::Gradient::NonMaxSobelSuppression ( const Image< D > & img )

Returns the sobel magnitude image with non-maximum intensities suppressed.

Date: 2013/Aug/08

Parameters

img	Input image.

Returns: Sobel magnitude image with non-maximum intensities suppressed.

Warning: none.

template<class D >

Image< D > Bial::Gradient::NonMaxSobelSuppression	(	const Image< D > &	magnitude,
		const Image< int > &	direction
	)

Returns the sobel magnitude image with non-maximum intensities suppressed.

Date: 2013/Aug/08

Parameters

magnitude	Sobel magnitude image.
direction	Sobel magnitude direction.

Returns: Sobel magnitude image with non-maximum intensities suppressed.

Warning: none.

template<class D >

void Bial::Gradient::RemoveBifurcation ( Image< D > & canny )

Removes points of bifurcation. It tries to keep straighter lines when possible.

Date: 2015/Jul/08

Parameters

canny A Canny gradient image.

Returns: none.

Warning: Only implemented for 2D images.

template<class D >

void Bial::Gradient::RemoveRedundantPoint ( Image< D > & canny )

Removes redundant points from components leaving the minimal width components.

Date: 2015/Jul/08

Parameters

canny A Canny gradient image.

Returns: none.

Warning: Only implemented for 2D images.

template<class D >

void Bial::Gradient::RemoveSmallSegments	(	Image< D > &	canny,
		size_t	min_seg_size
	)

Removes the edges that are smaller than the given size.

Date: 2015/Apr/29

Parameters

canny	A Canny gradient image.
min_seg_size	Minimum size of a segment.

Returns: none.

Warning: none.

template<class D >

Image< D > Bial::Gradient::ScaleCanny	(	const Image< D > &	img,
		size_t	window_scale,
		float	lower_threshold,
		float	higher_threshold
	)

Returns a scaled Canny edge detection image. Lower and higher threshold as frequencies from 0.0 to 1.0.

Date: 2014/Jun/09

Parameters

img	Input image.
window_scale	Scale by which image is divided to generate the window.
lower_threshold	Lower hysteresis threshold.
higher_threshold	Higher hysteresis threshold.

Returns: Scaled Canny edge detection image.

Warning: none.

template<class D >

void Bial::Gradient::Sobel	(	const Image< D > &	img,
		Image< D > *	magnitude = `nullptr`,
		Image< int > *	direction = `nullptr`
	)

Returns the directional Sobel filter over image.

Date: 2013/Nov/22

Parameters

img	Input image.
magnitude	Sobel magnitude image.
direction	Output Sobel direction image.

Returns: Directional Sobel filter over image. The directions are coded: 2D and 3D: No-dir:0 E:1 W:2 S:4 SE:5 SW:6 N:8 NE:9 NW:10 3D: U:16 UE:17 UW:18 US:20 NSE:21 USW:22 UN:24 UNE:25 UNW:26 3D: D:32 DE:33 DW:34 DS:36 DSE:37 DSW:38 DN:40 DNE:41 DNW:42

Warning: none.

template<class D >

Image< D > Bial::Gradient::SubScaleCanny	(	const Image< D > &	img,
		const Image< D > &	base_grad,
		size_t	window_scale,
		float	lower_threshold
	)

Returns a sub scaled Canny edge detection image. Lower threshold as a frequency from 0.0 to 1.0.

Date: 2015/Jan/27

Parameters

img	Input image.
base_grad	Gradient from higher levels with greater windows.
window_scale	Scale by which image is divided to generate the window.
lower_threshold	Lower hysteresis threshold.

Returns: Sub scaled Canny edge detection image.

Warning: none.

Functions

Detailed Description

Function Documentation