lti::csPresegmentation::parameters Class Reference

the parameters for the class csPresegmentation More...

#include <ltiCsPresegmentation.h>

Inheritance diagram for lti::csPresegmentation::parameters:

[legend]

Collaboration diagram for lti::csPresegmentation::parameters:

[legend]

List of all members.

Public Types
enum	eFilterType { Nothing = 0, Median = 1, KNearest = 2 }
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)
Public Attributes
int	smoothingKernelSize
kMColorQuantization::parameters	quantParameters
bool	useAlwaysNewPalette
int	borderSize
int	borderParts
bool	forceBorderToBackground
bool	labelObjects
float	backgroundTolerance
eFilterType	filterType
Static Public Attributes
static const int	All
static const int	Top
static const int	Bottom
static const int	Left
static const int	Right

---

Detailed Description

the parameters for the class csPresegmentation

---

Member Enumeration Documentation

enum lti::csPresegmentation::parameters::eFilterType

type to specify the smoothing filter to be used

Enumerator:

Nothing	No smoothing filter should be used.
Median	Median Filter.
KNearest	K-Nearest-Neighbour Filter.

---

Constructor & Destructor Documentation

lti::csPresegmentation::parameters::parameters

(

)

default constructor

Reimplemented from lti::segmentation::parameters.

lti::csPresegmentation::parameters::parameters

(

const parameters &

other

)

copy constructor

Parameters:

other

the parameters object to be copied

Reimplemented from lti::segmentation::parameters.

lti::csPresegmentation::parameters::~parameters

(

)

[virtual]

destructor

Reimplemented from lti::segmentation::parameters.

---

Member Function Documentation

virtual functor::parameters* lti::csPresegmentation::parameters::clone

(

)

const [virtual]

returns a pointer to a clone of the parameters

Reimplemented from lti::segmentation::parameters.

parameters& lti::csPresegmentation::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

Reimplemented from lti::segmentation::parameters.

const char* lti::csPresegmentation::parameters::getTypeName

(

)

const [virtual]

returns name of this type

Reimplemented from lti::segmentation::parameters.

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

read the parameters from 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::csPresegmentation::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

const int lti::csPresegmentation::parameters::All [static]

constant to define all border elements

float lti::csPresegmentation::parameters::backgroundTolerance

Tolerance for background colors.

Mean and Variance values will be computed for each RGB-component of all palette entries that are candidates to be background (at the beginning all entries in the palette are background candidates). Let t be this tolerance value. All entries in the palette, which are candidates to be background must satisfy for each color component c in RGB: in order to remain as background candidates.

Default value: 9;

int lti::csPresegmentation::parameters::borderParts

Decides which parts of the border should be considered for the statistics.

You can use combinations of constants Top, Bottom, Left and Right or All to produce the wished results.

Default value: lti::csPresegmentation::parameters::All

Examples:

 lti::csPresegmentation::parameters param;

 // to consider all but the bottom of the image:
 param.borderParts = lti::csPresegmentation::parameters::All &
                     ~lti::csPresegmentation::parameters::Bottom;


 // to consider only the left and right borders
 param.borderParts = lti::csPresegmentation::parameters::Left |
                     lti::csPresegmentation::parameters::Right;

Bottom and Top parts have the whole width of the image. Left and Right parts have a height equal to the height of the image minus twice the borderSize:

 _____________________
 |        Top        |
 |___________________|
 | L|             |R |
 | e|             |i |
 | f|             |g |
 | t|             |h |
 |  |             |t |
 |__|_____________|__|
 |      Bottom       |
 |___________________|

int lti::csPresegmentation::parameters::borderSize

Maximal size of the border of the image to be considered to contain almost background pixels.

The real border will have the smallest number between this parameter and one half of the rows or columns of the image being used.

Default value: 16

const int lti::csPresegmentation::parameters::Bottom [static]

constant to define the bottom border element

eFilterType lti::csPresegmentation::parameters::filterType

Smoothing filter to be used.

Default: Median

bool lti::csPresegmentation::parameters::forceBorderToBackground

If true, all color entries found withing the given border will be assumed to be background, undependently of their values.

If false, only the most representative colors in the border will be assumed to be background.

Default value: false;

bool lti::csPresegmentation::parameters::labelObjects

If true, the value in the segmentation mask for each objekt will correspond to the detected color palette entry in the color quantization.

If false, the value in the mask for the non-background pixels will be 255

Default value: false;

const int lti::csPresegmentation::parameters::Left [static]

constant to define the left border element

kMColorQuantization::parameters lti::csPresegmentation::parameters::quantParameters

Parameters for the k-means color quantization.

Default values: quantParameters.numberOfColors=12 quantParameters.thresholdDeltaPalette = 0.5f quantParameters.maximalNumberOfIterations = 50

const int lti::csPresegmentation::parameters::Right [static]

constant to define the right border element

int lti::csPresegmentation::parameters::smoothingKernelSize

Size of the median filter kernel or k-Nearest-Neighbor filter used to smooth the quantization mask.

Default value: 5

const int lti::csPresegmentation::parameters::Top [static]

constant to define the top border element

bool lti::csPresegmentation::parameters::useAlwaysNewPalette

If false, the quantization algorithms uses the last palette as preinitialization, to speed up the segmentation of several images showing the same object from different perspectives.

If true, the last quantization will be always ignored.

Default value: false

---

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

• ltiCsPresegmentation.h