Classes
class	UniqueNumber

Enumerations
enum	{ RET_OK = 0x00000001UL, RET_FAILED = 0x00000002UL, RET_WARNING = 0x00000004UL }

Functions
template<typename Distance >
void	ClusteringKMeanspp (int k, cv::Mat &dataset, int indices, int indices_length, int centers, int &centers_length, int numLocalTries=1)

unsigned long	ConvertToShowImage (const cv::Mat &source, cv::Mat &dest, int channel=1, double min=-1, double max=-1)

unsigned long	EvaluatePolynomial (double x, int degree, double coefficients, double y)

unsigned long	FilterByAmplitude (cv::Mat &xyzImage, const cv::Mat &greyImage, cv::Mat mask, cv::Mat maskColor, float minMaskThresh, float maxMaskThresh)

unsigned long	FilterSpeckles (cv::Mat &img, int maxSpeckleSize, double _maxDiff, cv::Mat &_buf)
	Description. More...

unsigned long	FilterTearOffEdges (cv::Mat &xyzImage, cv::Mat *mask, float piHalfFraction=6)

cv::Vec3b	GrayColorMap (double value, double min, double max)

unsigned long	LoadMat (cv::Mat &mat, std::string filename, int type=CV_32F)

unsigned long	MaskImage (const cv::Mat &source, cv::Mat &dest, const cv::Mat &mask, cv::Mat &destMask, float minMaskThresh=0, float maxMaskTresh=20000, int sourceChannel=1, double sourceMin=-1, double sourceMax=-1)

unsigned long	SaveMat (cv::Mat &mat, std::string filename, int type=CV_32F)

template<class T >
std::vector< T >	takeRandomN (const std::vector< T > &v, int n)

cv::Mat	vstack (const std::vector< cv::Mat > &mats)

Enumeration Type Documentation

anonymous enum

An enum for the return values. This enum describes possible return values that are used to return failure or success.

Enumerator
RET_OK	Everythings OK.
RET_FAILED	Something went dramatically wrong.
RET_WARNING	Something went wrong.

Definition at line 25 of file GlobalDefines.h.

Function Documentation

template<typename Distance >

void ipa_Utils::ClusteringKMeanspp	(	int	k,
		cv::Mat &	dataset,
		int *	indices,
		int	indices_length,
		int *	centers,
		int &	centers_length,
		int	numLocalTries = `1`
	)

Original implementation from OpenCV, hierarchical_clustering_index.h Chooses the initial centers in the k-means using the algorithm proposed in the KMeans++ paper: Arthur, David; Vassilvitskii, Sergei - k-means++: The Advantages of Careful Seeding

Definition at line 113 of file VisionUtils.h.

unsigned long ipa_Utils::ConvertToShowImage	(	const cv::Mat &	source,
		cv::Mat &	dest,
		int	channel = `1`,
		double	min = `-1`,
		double	max = `-1`
	)

Function to convert a 32 bit, n channel image into a eight bit, 1 channel image.

Parameters

source	The 32 bit, n channel source image
dest	The resulting 8 bit, 1 channel image
channel	The channel of the source image to process
min	Minimal value that is converted, smaller values are clipped to min
max	Maximal value that is converted, bigger values are clipped to max

Returns: Return code

Calculate minmal and maximal value within the specified image channel

Definition at line 264 of file VisionUtils.cpp.

unsigned long ipa_Utils::EvaluatePolynomial	(	double	x,
		int	degree,
		double *	coefficients,
		double *	y
	)

Evaluates a polynomial, calculated with FitPolynomial

Parameters

x	The function value
degree	The degree of the polynomial
coefficients	Pointer to an array of the `degree`+1 coefficients fo the polynomial
y	The resulting function value

Returns: Return code

Definition at line 63 of file VisionUtils.cpp.

unsigned long ipa_Utils::FilterByAmplitude	(	cv::Mat &	xyzImage,
		const cv::Mat &	greyImage,
		cv::Mat *	mask,
		cv::Mat *	maskColor,
		float	minMaskThresh,
		float	maxMaskThresh
	)

Filters a 3D range image with help of the amplitude image

Parameters

xyzImage	A 3 channel, 32bit image, containing the xyz-data, filtered values are set to 0.
greyImage	A 3 channel, 32bit image, containing the amplitude data
mask	The resulting mask image (32F1).
maskColor	The resulting color mask. Pixels are set to different colors if they are filtered or not.
minMaskThresh	Lower border for filtering.
maxMaskThresh	Upper border for filtering.

build color mask from amplitude values

Definition at line 395 of file VisionUtils.cpp.

unsigned long ipa_Utils::FilterSpeckles	(	cv::Mat &	img,
		int	maxSpeckleSize,
		double	_maxDiff,
		cv::Mat &	_buf
	)

Description.

Definition at line 703 of file VisionUtils.cpp.

unsigned long ipa_Utils::FilterTearOffEdges	(	cv::Mat &	xyzImage,
		cv::Mat *	mask,
		float	piHalfFraction = `6`
	)

Filters tear-off edges from a 3D range image. All tear off edges are masked with a value of 255 in mask image. All other points are 0.

Parameters

xyzImage	A 3 channel, 32bit image, containing the xyz-data, filtered values are set to 0.
mask	The resulting image mask
piHalfFraction	Angles between (PI/2)/'piHalfFraction' and (2*PI)-(PI/2)/'piHalfFraction' are not discarded

Returns: Return code

Check if destination image has been initialized correctly

Extract four surrounding neighbor vectors that have a non zero mask value

x x o x x

Counte the amount of times, we satisfy the thresholds

Vector Middle (must exist)

Vector Left

Vector Right

Vector Up

Vector Down

Mask value if angle exceeded threshold too often

Definition at line 482 of file VisionUtils.cpp.

cv::Vec3b ipa_Utils::GrayColorMap	(	double	value,
		double	min,
		double	max
	)

Returns mat as HSV or gray image

Parameters

value	Value to convert
min	Minimum for scaling
max	Maximum for scaling

Returns: Vector containing RGB values

Definition at line 808 of file VisionUtils.cpp.

unsigned long ipa_Utils::LoadMat	(	cv::Mat &	mat,
		std::string	filename,
		int	type = `CV_32F`
	)

Load OpenCV matrix in binary format.

Parameters

mat	The OpenCV mat data structure
filename	The filename

Returns: Return code

Definition at line 865 of file VisionUtils.cpp.

unsigned long ipa_Utils::MaskImage	(	const cv::Mat &	source,
		cv::Mat &	dest,
		const cv::Mat &	mask,
		cv::Mat &	destMask,
		float	minMaskThresh = `0`,
		float	maxMaskTresh = `20000`,
		int	sourceChannel = `1`,
		double	sourceMin = `-1`,
		double	sourceMax = `-1`
	)

Function to mask image source with image mask. All Pixels in source will be set to 0 if the corresponding pixel in mask is outside the interval [minMaskThresh, maxMaskTresh]. Additionally the source image values can be bounded to [sourceMin, sourceMax].

Parameters

source	The source image.
dest	The destination image.
mask	The image used as mask for the source image.
destMask	The resulting destination mask image (32F1).
minMaskThresh	Lower border for masking.
maxMaskThresh	Upper border for masking.
sourceChannel	Channel to be bounded in source image.
sourceMin	Lower border for bounding source image.
sourceMax	Upper border for bounding source image.

Check if destination image has been initialized correctly

Check if mask image has been initialized correctly

Calculate minmal and maximal value within the specified image sourceChannel Channel must be within [1, source->nChannels]

Definition at line 75 of file VisionUtils.cpp.

unsigned long ipa_Utils::SaveMat	(	cv::Mat &	mat,
		std::string	filename,
		int	type = `CV_32F`
	)

Save OpenCV matrix in binary format.

Parameters

mat	The OpenCV mat data structure
filename	The filename

Returns: Return code

Definition at line 817 of file VisionUtils.cpp.

template<class T >

std::vector<T> ipa_Utils::takeRandomN	(	const std::vector< T > &	v,
		int	n
	)

Definition at line 196 of file VisionUtils.h.

cv::Mat ipa_Utils::vstack ( const std::vector< cv::Mat > & mats )

Combines different matrices row wise, so that the number of columns remains the same and the number of rows increases.

Returns: The stacked matrix

Definition at line 26 of file VisionUtils.cpp.

Classes

Enumerations

Functions

Enumeration Type Documentation

Function Documentation