PUBLIC

RegClassColC::RegClassColC(void) RegClassColC::RegClassColC(IntT) RegClassColC::RegClassColC(ValT) RegClassColC::Initialise(void) RegClassColC::Contains(const PixelC &,const ValT &) const RegClassColC::Distance(const PixelC &,const ValT &) const RegClassColC::Include(const PixelC &,const ValT &) RegClassColC::Remove(const PixelC &,const ValT &) RegClassColC::Counter(void) RegClassColC::Stat(void) RegClassColC::Save(ostream &) const

RegClassColC<class ValT,class StatT>
Class to determine whether a pixel belongs to a region based on its colour.
include	"amma/Region/RegClassCol.hh"
User Level:	Default
Library:	IPRegion
Example:	exSeg.cc
Section:	Image.Image Processing.Region.Homogeneity Criteria
In Scope:	std

Comments:
At the moment, the only statistic which can be calculated for the region is the mean. The colour representation can only be 3-dimensional. StatT can be any pixel representation class which is inherited from Point3dC. ValT can be any pixel representation which can be used to construct an object of class StatT.

int count;

StatT total;

StatT mean;

StatT threshold;

RegClassColC()

default constructor

RegClassColC(IntT thr)

constructor

RegClassColC(ValT thr)

constructor

void Initialise()

Resets the statistics for the region

BooleanT Contains(const PixelC & pxl,const ValT & val) const

Return TRUE if pxl belongs to the region

RealT Distance(const PixelC & pxl,const ValT & val) const

Distance to the mean

void Include(const PixelC & pxl,const ValT & val)

Update the statistics of the region for the included pixel

void Remove(const PixelC & pxl,const ValT & val)

Update the statistics of the region for the stolen pixel

IntT Counter()

the number of the member pixels within this cluster

StatT Stat()

Return the statistics about the region

BooleanT Save(ostream & out) const