cv::stereo Namespace Reference

Classes
struct	MatchQuasiDense
struct	PropagationParameters
class	QuasiDenseStereo
	Class containing the methods needed for Quasi Dense Stereo computation. More...
class	StereoBinaryBM
	Class for computing stereo correspondence using the block matching algorithm, introduced and contributed to OpenCV by K. Konolige. More...
class	StereoBinarySGBM
	The class implements the modified H. Hirschmuller algorithm [126] that differs from the original one as follows: More...

Enumerations
enum	{   CV_DENSE_CENSUS ,   CV_SPARSE_CENSUS ,   CV_CS_CENSUS ,   CV_MODIFIED_CS_CENSUS ,   CV_MODIFIED_CENSUS_TRANSFORM ,   CV_MEAN_VARIATION ,   CV_STAR_KERNEL }
	types of supported kernels More...
enum	{   CV_SPECKLE_REMOVAL_ALGORITHM ,   CV_SPECKLE_REMOVAL_AVG_ALGORITHM }
	speckle removal algorithms. These algorithms have the purpose of removing small regions More...
enum	{   CV_QUADRATIC_INTERPOLATION ,   CV_SIMETRICV_INTERPOLATION }
	subpixel interpolationm methods for disparities. More...

Functions
void	censusTransform (const Mat &image1, const Mat &image2, int kernelSize, Mat &dist1, Mat &dist2, const int type)
void	censusTransform (const Mat &image1, int kernelSize, Mat &dist1, const int type)
	single image census transform
void	modifiedCensusTransform (const Mat &img1, const Mat &img2, int kernelSize, Mat &dist1, Mat &dist2, const int type, int t=0, const Mat &integralImage1=Mat(), const Mat &integralImage2=Mat())
void	modifiedCensusTransform (const Mat &img1, int kernelSize, Mat &dist, const int type, int t=0, const Mat &integralImage=Mat())
	single version of modified census transform descriptor
void	starCensusTransform (const Mat &img1, const Mat &img2, int kernelSize, Mat &dist1, Mat &dist2)
	in a 9x9 kernel only certain positions are choosen
void	starCensusTransform (const Mat &img1, int kernelSize, Mat &dist)
	single image version of star kernel
void	symetricCensusTransform (const Mat &img1, const Mat &img2, int kernelSize, Mat &dist1, Mat &dist2, const int type)
void	symetricCensusTransform (const Mat &img1, int kernelSize, Mat &dist1, const int type)
	single version of census transform

Enumeration Type Documentation

anonymous enum

anonymous enum

types of supported kernels

Enumerator
CV_DENSE_CENSUS  Python: cv.stereo.CV_DENSE_CENSUS
CV_SPARSE_CENSUS  Python: cv.stereo.CV_SPARSE_CENSUS
CV_CS_CENSUS  Python: cv.stereo.CV_CS_CENSUS
CV_MODIFIED_CS_CENSUS  Python: cv.stereo.CV_MODIFIED_CS_CENSUS
CV_MODIFIED_CENSUS_TRANSFORM  Python: cv.stereo.CV_MODIFIED_CENSUS_TRANSFORM
CV_MEAN_VARIATION  Python: cv.stereo.CV_MEAN_VARIATION
CV_STAR_KERNEL  Python: cv.stereo.CV_STAR_KERNEL

Function Documentation

censusTransform() [1/2]

void cv::stereo::censusTransform	(	const Mat &	image1,
		const Mat &	image2,
		int	kernelSize,
		Mat &	dist1,
		Mat &	dist2,
		const int	type
	)

Two variations of census applied on input images Implementation of a census transform which is taking into account just the some pixels from the census kernel thus allowing for larger block sizes

censusTransform() [2/2]

void cv::stereo::censusTransform	(	const Mat &	image1,
		int	kernelSize,
		Mat &	dist1,
		const int	type
	)

single image census transform

modifiedCensusTransform() [1/2]

void cv::stereo::modifiedCensusTransform	(	const Mat &	img1,
		const Mat &	img2,
		int	kernelSize,
		Mat &	dist1,
		Mat &	dist2,
		const int	type,
		int	t = 0,
		const Mat &	integralImage1 = Mat(),
		const Mat &	integralImage2 = Mat()
	)

STANDARD_MCT - Modified census which is memorizing for each pixel 2 bits and includes a tolerance to the pixel comparison MCT_MEAN_VARIATION - Implementation of a modified census transform which is also taking into account the variation to the mean of the window not just the center pixel

modifiedCensusTransform() [2/2]

void cv::stereo::modifiedCensusTransform	(	const Mat &	img1,
		int	kernelSize,
		Mat &	dist,
		const int	type,
		int	t = 0,
		const Mat &	integralImage = Mat()
	)

single version of modified census transform descriptor

starCensusTransform() [1/2]

void cv::stereo::starCensusTransform	(	const Mat &	img1,
		const Mat &	img2,
		int	kernelSize,
		Mat &	dist1,
		Mat &	dist2
	)

in a 9x9 kernel only certain positions are choosen

starCensusTransform() [2/2]

void cv::stereo::starCensusTransform	(	const Mat &	img1,
		int	kernelSize,
		Mat &	dist
	)

single image version of star kernel

symetricCensusTransform() [1/2]

void cv::stereo::symetricCensusTransform	(	const Mat &	img1,
		const Mat &	img2,
		int	kernelSize,
		Mat &	dist1,
		Mat &	dist2,
		const int	type
	)

The classical center symetric census A modified version of cs census which is comparing a pixel with its correspondent after the center

symetricCensusTransform() [2/2]

void cv::stereo::symetricCensusTransform	(	const Mat &	img1,
		int	kernelSize,
		Mat &	dist1,
		const int	type
	)

single version of census transform