latest version v1.9 - last update 24 Nov 2005

lti::meanShiftSegmentation::parameters Class Reference

the parameters for the class meanShiftSegmentation More...

#include <ltiMeanShiftSegmentation.h>

Inheritance diagram for lti::meanShiftSegmentation::parameters:

[legend]Collaboration diagram for lti::meanShiftSegmentation::parameters:

[legend]List of all members.

Parameters for the old algorithm
enum	{ Quantization = 0, Oversegmentation = 1, Undersegmentation = 2 }
int	option
std::vector< rectangle >	rects
int	maxTrial
int	trial2converge
double	classThreshold [3]
int	maxTrialRandomColor
int	minRegionSize
float	rectRadius [3]
float	autoRadius [3]
float	minVar
Parameters for the new algorithm
enum	eSpeedUpType { NoSpeedup, MediumSpeedup, HighSpeedup }
bool	multivariateNormalKernel
eSpeedUpType	speedup
double	sigmaS
double	sigmaR
double	maxNeighbourColorDistance
double	thresholdConverged
bool	classicAlgorithm
Public Member Functions
	parameters ()
	parameters (const parameters &other)
	~parameters ()
const char *	getTypeName () const
parameters &	copy (const parameters &other)
parameters &	operator= (const parameters &other)
virtual functor::parameters *	clone () const
virtual bool	write (ioHandler &handler, const bool complete=true) const
virtual bool	read (ioHandler &handler, const bool complete=true)

---

Detailed Description

the parameters for the class meanShiftSegmentation

---

Member Enumeration Documentation

anonymous enum

Constants to specify degree of segmentation. Enumerator: Quantization  Constant value to specify the quantization option. Oversegmentation  constant value to specify the oversegmentation option. Undersegmentation  constant value to specify the undersegmentation option.

enum lti::meanShiftSegmentation::parameters::eSpeedUpType

Three types of speed-up techniques. Enumerator: NoSpeedup  Filters the image applying mean shift to each point. Advantage: most accurate Disadvantage : time expensive MediumSpeedup  Filters the image using previous mode information to avoid re-applying mean shift to some data points (is default). Advantage: maintains high level of accuracy, large speed up compared to non-optimized version Disadvantage: possibly not as accurate as non-optimized version HighSpeedup  filters the image using previous mode information and window traversals to avoid re-applying mean shift to some data points advantage: huge speed up maintains accuracy good enough for segmentation disadvantage: not as accurate as previous filters

---

Constructor & Destructor Documentation

lti::meanShiftSegmentation::parameters::parameters (   )

default constructor Reimplemented from lti::segmentation::parameters.

lti::meanShiftSegmentation::parameters::parameters (  const parameters &  other  )

copy constructor Parameters: other  the parameters object to be copied

lti::meanShiftSegmentation::parameters::~parameters (   )  [virtual]

destructor Reimplemented from lti::segmentation::parameters.

---

Member Function Documentation

virtual functor::parameters* lti::meanShiftSegmentation::parameters::clone (   )  const [virtual]

returns a pointer to a clone of the parameters Reimplemented from lti::segmentation::parameters.

parameters& lti::meanShiftSegmentation::parameters::copy (  const parameters &  other  )

copy the contents of a parameters object Parameters: other  the parameters object to be copied Returns: a reference to this parameters object

const char* lti::meanShiftSegmentation::parameters::getTypeName (   )  const [virtual]

returns name of this type Reimplemented from lti::segmentation::parameters.

parameters& lti::meanShiftSegmentation::parameters::operator= (  const parameters &  other  )

copy the contents of a parameters object Parameters: other  the parameters object to be copied Returns: a reference to this parameters object

virtual bool lti::meanShiftSegmentation::parameters::read (  ioHandler &  handler, const bool  complete = true )  [virtual]

write the parameters in the given ioHandler Parameters: handler  the ioHandler to be used complete  if true (the default) the enclosing begin/end will be also written, otherwise only the data block will be written. Returns: true if write was successful Reimplemented from lti::segmentation::parameters.

virtual bool lti::meanShiftSegmentation::parameters::write (  ioHandler &  handler, const bool  complete = true )  const [virtual]

write the parameters in the given ioHandler Parameters: handler  the ioHandler to be used complete  if true (the default) the enclosing begin/end will be also written, otherwise only the data block will be written. Returns: true if write was successful Reimplemented from lti::segmentation::parameters.

---

Member Data Documentation

float lti::meanShiftSegmentation::parameters::autoRadius[3]

Multiplication factor for the colorRadius(option). Use in auto-segmentation mode (see rects). Increase value if too many colors are found, because of a low variance(image) Each element of the array corresponds to one of the possible options (see option parameter); Default value: {2.0, 3.0, 4.0} for Quantization, Over- and Undersegmentation respectively.

bool lti::meanShiftSegmentation::parameters::classicAlgorithm

Choose algorithm by this parameter. true: the old algorithm (for compatibility reasons still provided) false: the new implementation, like in EDISON. The classic algorithm uses the following parameters: option rects maxTrial trial2converge classThreshold maxTrialRandomColor minRegionSize rectRadius autoRadius minVar The new algorithm considers the following parameters; maxTrial speedup sigmaS sigmaR maxNeighbourColorDistance thresholdConverged Default value: true (classic algorithm is used)

double lti::meanShiftSegmentation::parameters::classThreshold[3]

How many pixel in promille of the image must have a colorClass. Each element of the array corresponds to one of the possible options (see option parameter) Default value: {2.5, 5.0, 10.0} for Quantization, Over- and Undersegmentation respectively.

double lti::meanShiftSegmentation::parameters::maxNeighbourColorDistance

Regions having a color difference less than this parameter are joined together (by the method fuseRegions()). It should be smaller than sigmaR. Default value: 3

int lti::meanShiftSegmentation::parameters::maxTrial

Maximal number of trials for choosing a valid representing color. Default: 10

int lti::meanShiftSegmentation::parameters::maxTrialRandomColor

Number of trials to pick up randomly a suitable color in the image(rect). Default: 25

int lti::meanShiftSegmentation::parameters::minRegionSize

Set the minSize (in Pixel) a region must have. default: 15

float lti::meanShiftSegmentation::parameters::minVar

Set the lower limit of a multiplication factor v for the colorRadius. max( minVar , v ) The factor v = sqrt((var.l+var.u+var.v)/100) depends on the variance of the image in luv. For images with an homogeneous background and a small object, this value could be increased to get better results (for example with 1.0). Default value: 0.0 (no influence on factor v)

bool lti::meanShiftSegmentation::parameters::multivariateNormalKernel

Three types of speed-up techniques.

int lti::meanShiftSegmentation::parameters::option

Option: 0 Quantization 1 Oversegmentation 2 Undersegmentation. You should use the given constants instead of the magic numbers... Default: Undersegmentation

float lti::meanShiftSegmentation::parameters::rectRadius[3]

Multiplication factor for the colorRadius (variance in image). Only used in auto-segmentation mode (see rects). Each element of the array corresponds to one of the possible options (see option parameter); Default value: {8.0, 6.0, 4.0} for Quantization, Over- and Undersegmentation respectively.

std::vector<rectangle> lti::meanShiftSegmentation::parameters::rects

Number of rectangles, in which the segmenter looks for a color. If the first rectangle is (0,0,0,0) then an auto-segmentation mode is activated and colors are taken from the whole image. Otherwise the colors will be taken from the given image region. The number of rectangles specifies the maximal possible number of colors for the segmented/quantized image. If you give only one rectangle with (0,0,0,0) or an empty vector, then 50 elements will be assumed. Default value: vector with 1 element (0,0,0,0)

double lti::meanShiftSegmentation::parameters::sigmaR

The range radius of the mean shift sphere. (the radius in color space) Higher values result in less regions Default value: 5

double lti::meanShiftSegmentation::parameters::sigmaS

The spatial radius of the mean shift sphere. (the radius in grid space) Higher values cause longer computation times, and smoother region boundaries. Default value: 5

eSpeedUpType lti::meanShiftSegmentation::parameters::speedup

Higher speedup level causes loss of acuracy. Default: MediumSpeedup

double lti::meanShiftSegmentation::parameters::thresholdConverged

If the magnitude of the mean-shift vector is under this threshold, it is considered as converged. Default value: 0.1

int lti::meanShiftSegmentation::parameters::trial2converge

Number of trials for the representing color to converge. Default: 15 trials

---

The documentation for this class was generated from the following file:

• ltiMeanShiftSegmentation.h

---