icl::cv::RegionGrower Class Reference

class for region growing on images and DataSegments (e.g. poincloud xyzh) More...

#include <RegionGrower.h>

List of all members.

Classes
struct	EqualThreshold
struct	Float4EuclideanDistance
struct	RegionGrowerDataAccessor
struct	RegionGrowerDataAccessor< core::DataSegment< float, 4 >, float, 4 >
struct	RegionGrowerDataAccessor< core::Img8u, icl8u, 1 >
struct	RegionGrowerDataAccessor< core::Img8u, icl8u, 3 >
Public Member Functions
template<class Criterion >
const core::Img32s &	apply (const core::Img8u &image, Criterion crit, core::Img8u *initialMask=0, const unsigned int minSize=0, const unsigned int startID=1)
	Applies the region growing on an input image with a growing criterion.
template<class Criterion >
const core::Img32s &	apply (const core::DataSegment< float, 4 > &dataseg, Criterion crit, core::Img8u *initialMask=0, const unsigned int minSize=0, const unsigned int startID=1)
	Applies the region growing on an input data segment with a growing criterion.
const core::Img32s &	applyFloat4EuclideanDistance (const core::DataSegment< float, 4 > &dataseg, core::Img8u mask, const int threshold, const unsigned int minSize=0, const unsigned int startID=1)
	Applies the region growing on an input data segment with euclidean distance criterion.
const core::Img32s &	applyEqualThreshold (const core::Img8u &image, core::Img8u mask, const int threshold, const unsigned int minSize=0, const unsigned int startID=1)
	Applies the region growing on an input image with value-equals-threshold criterion.
std::vector< std::vector< int > >	getRegions ()
	Returns a vector of regions containing the image IDs. This is an additional representation of the result.
Private Member Functions
template<class T , class DataT , int DIM, class Criterion >
void	region_grow (const T &data, core::Img8u &mask, core::Img32s &result, Criterion crit, const unsigned int minSize, const unsigned int startID=1)
Static Private Member Functions
template<class T , class DataT , int DIM, class Criterion >
static void	flood_fill (const RegionGrowerDataAccessor< T, DataT, DIM > &a, int xStart, int yStart, core::Channel8u &processed, Criterion crit, std::vector< int > &result, core::Channel32s &result2, int id)
Private Attributes
core::Img8u	mask
core::Img32s	result
std::vector< std::vector< int > >	regions

---

Detailed Description

class for region growing on images and DataSegments (e.g. poincloud xyzh)

The RegionGrower class is designed as template applying a growing criterion to given input data. A mask defines the points for processing (e.g. a region of interest).

---

Member Function Documentation

template<class Criterion >

const core::Img32s& icl::cv::RegionGrower::apply	(	const core::Img8u &	image,
		Criterion	crit,
		core::Img8u *	initialMask = 0,
		const unsigned int	minSize = 0,
		const unsigned int	startID = 1
	)		[inline]

Applies the region growing on an input image with a growing criterion.

Parameters:

image	the input image for region growing
crit	the region growing criterion
initialMask	the initial mask (e.g. ROI)
minSize	the minimum size of regions (smaller regions are removed)
startID	the start id for the result label image

Returns:

the result label image

template<class Criterion >

const core::Img32s& icl::cv::RegionGrower::apply	(	const core::DataSegment< float, 4 > &	dataseg,
		Criterion	crit,
		core::Img8u *	initialMask = 0,
		const unsigned int	minSize = 0,
		const unsigned int	startID = 1
	)		[inline]

Applies the region growing on an input data segment with a growing criterion.

Parameters:

dataseg	the input data segment for region growing
crit	the region growing criterion
initialMask	the initial mask (e.g. ROI)
minSize	the minimum size of regions (smaller regions are removed)
startID	the start id for the result label image

Returns:

the result label image

const core::Img32s& icl::cv::RegionGrower::applyEqualThreshold	(	const core::Img8u &	image,
		core::Img8u	mask,
		const int	threshold,
		const unsigned int	minSize = 0,
		const unsigned int	startID = 1
	)		[inline]

Applies the region growing on an input image with value-equals-threshold criterion.

Parameters:

image	the input image for region growing
mask	the initial mask (e.g. ROI)
threshold	the equals-to-value (growing criterion)
minSize	the minimum size of regions (smaller regions are removed)
startID	the start id for the result label image

Returns:

the result label image

const core::Img32s& icl::cv::RegionGrower::applyFloat4EuclideanDistance	(	const core::DataSegment< float, 4 > &	dataseg,
		core::Img8u	mask,
		const int	threshold,
		const unsigned int	minSize = 0,
		const unsigned int	startID = 1
	)		[inline]

Applies the region growing on an input data segment with euclidean distance criterion.

Parameters:

dataseg	the input data segment for region growing
mask	the initial mask (e.g. ROI)
threshold	the maximum euclidean distance
minSize	the minimum size of regions (smaller regions are removed)
startID	the start id for the result label image

Returns:

the result label image

template<class T , class DataT , int DIM, class Criterion >

void icl::cv::RegionGrower::flood_fill	(	const RegionGrowerDataAccessor< T, DataT, DIM > &	a,
		int	xStart,
		int	yStart,
		core::Channel8u &	processed,
		Criterion	crit,
		std::vector< int > &	result,
		core::Channel32s &	result2,
		int	id
	)		[static, private]

std::vector<std::vector<int> > icl::cv::RegionGrower::getRegions

(

)

[inline]

Returns a vector of regions containing the image IDs. This is an additional representation of the result.

Returns:

the vector of regions with the image IDs.

template<class T , class DataT , int DIM, class Criterion >

void icl::cv::RegionGrower::region_grow	(	const T &	data,
		core::Img8u &	mask,
		core::Img32s &	result,
		Criterion	crit,
		const unsigned int	minSize,
		const unsigned int	startID = 1
	)		[inline, private]

---

Member Data Documentation

core::Img8u icl::cv::RegionGrower::mask [private]

std::vector<std::vector<int> > icl::cv::RegionGrower::regions [private]

core::Img32s icl::cv::RegionGrower::result [private]

---

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

• /home/jenkins/workspace/icl-manual-trunk/icl-manual/ICLCV/src/ICLCV/RegionGrower.h