#include <itkWatershedBoundaryResolver.h>

Inheritance diagram for itk::watershed::BoundaryResolver< TPixelType, TDimension >:

Collaboration diagram for itk::watershed::BoundaryResolver< TPixelType, TDimension >:

Public Types
typedef Boundary< PixelType, TDimension >	BoundaryType
typedef DataObject::Pointer	DataObjectPointer
typedef ProcessObject::DataObjectPointerArraySizeType	DataObjectPointerArraySizeType
typedef EquivalencyTable	EquivalencyTableType
typedef TPixelType	PixelType
typedef Segmenter< Image < TPixelType, TDimension > >	SegmenterType
Public Member Functions
void	GenerateData ()
virtual DataObjectPointer	MakeOutput (DataObjectPointerArraySizeType idx)

void	SetBoundaryA (BoundaryType *bd)
BoundaryType::Pointer	GetBoundaryA ()

void	SetBoundaryB (BoundaryType *bd)
BoundaryType::Pointer	GetBoundaryB ()

virtual void	SetFace (unsigned short _arg)
virtual unsigned short	GetFace () const

void	SetEquivalencyTable (EquivalencyTableType::Pointer a)
EquivalencyTableType::Pointer	GetEquivalencyTable ()
Static Public Attributes
static const unsigned int	ImageDimension = TDimension
Protected Member Functions
	BoundaryResolver ()
	BoundaryResolver (const Self &)
void	GenerateOutputRequestedRegion (DataObject *output)
void	operator= (const Self &)
void	PrintSelf (std::ostream &os, Indent indent) const
virtual	~BoundaryResolver ()
Protected Attributes
unsigned short	m_Face
typedef BoundaryResolver	Self
typedef ProcessObject	Superclass
typedef SmartPointer< Self >	Pointer
typedef SmartPointer< const Self >	ConstPointer
static Pointer	New ()
virtual ::itk::LightObject::Pointer	CreateAnother (void) const
virtual const char *	GetNameOfClass () const

Detailed Description

template<class TPixelType, unsigned int TDimension>
class itk::watershed::BoundaryResolver< TPixelType, TDimension >

This filter implements a piece of the streaming watershed segmentation algorithm. It takes in pairs of itk::watershed::Boundary objects and connects the labeling of pixels across image chunk boundaries. Read the documentation found in itk::WatershedImageFilter and the other watershed segmentation component objects for more information.

A note on terminology in itk watershed segmentation code: For streamed segmentation of images in the watershed framework, the documentation refers to the complete data set at the ``image volume.'' The image volume is assumed to be partitioned into pieces referred to as image chunks. Each chunk is processed in sequence through the pipeline. The ``face'' of an image chunk is an N-1 dimensional boundary region of an N dimensional chunk (the planar faces of a cube, for example).

Input: The input to this filter is a pair of itk::watershed::Boundary pointers (BoundaryA and BoundaryB). The algorithm assumes that these Boundaries come from facing chunks in the image volume. The faces that align need to be specified in the parameters of the filter.

Output: This filter outputs a table of equivalencies among labels. See itk::EquivalencyTable for more information.

Parameters: The only parameters to set on this filter are the indicies of the faces that align in the boundary object inputs. See itk::Boundary for a description of how boundary faces are indexed.

See also:: itk::watershed::Boundary

Definition at line 64 of file itkWatershedBoundaryResolver.h.