#include <itkBinaryMorphologyImageFilter.h>

Inheritance diagram for itk::BinaryMorphologyImageFilter< TInputImage, TOutputImage, TKernel >:

Collaboration diagram for itk::BinaryMorphologyImageFilter< TInputImage, TOutputImage, TKernel >:

Public Types
typedef SmartPointer< const Self >	ConstPointer
typedef InputImageType::IndexType	IndexType
typedef InputImageType::IndexValueType	IndexValueType
typedef InputImageType::RegionType	InputImageRegionType
typedef TInputImage	InputImageType
typedef InputImageType::PixelType	InputPixelType
typedef NumericTraits < InputPixelType >::RealType	InputRealType
typedef InputImageType::SizeType	InputSizeType
typedef InputImageType::SizeValueType	InputSizeValueType
typedef KernelType::ConstIterator	KernelIteratorType
typedef TKernel	KernelType
typedef InputImageType::OffsetType	OffsetType
typedef OutputImageType::RegionType	OutputImageRegionType
typedef TOutputImage	OutputImageType
typedef OutputImageType::PixelType	OutputPixelType
typedef SmartPointer< Self >	Pointer
typedef BinaryMorphologyImageFilter	Self
typedef KernelImageFilter < InputImageType, OutputImageType, TKernel >	Superclass
Public Member Functions
virtual ::itk::LightObject::Pointer	CreateAnother (void) const
virtual OutputPixelType	GetBackgroundValue () const
virtual InputPixelType	GetForegroundValue () const
virtual const char *	GetNameOfClass () const
virtual void	SetBackgroundValue (OutputPixelType _arg)
virtual void	SetForegroundValue (InputPixelType _arg)
void	SetKernel (const KernelType &kernel)
	typedef (Concept::SameDimension< itkGetStaticConstMacro(InputImageDimension), itkGetStaticConstMacro(OutputImageDimension) >) ImageDimensionCheck

virtual void	SetBoundaryToForeground (bool _arg)
virtual const bool &	GetBoundaryToForeground ()
virtual void	BoundaryToForegroundOn ()
virtual void	BoundaryToForegroundOff ()
Static Public Member Functions
static Pointer	New ()
Static Public Attributes
static const unsigned int	KernelDimension = TKernel::NeighborhoodDimension

static const unsigned int	InputImageDimension = TInputImage::ImageDimension
static const unsigned int	OutputImageDimension = TOutputImage::ImageDimension
Protected Types
typedef ComponentVectorType::const_iterator	ComponentVectorConstIterator
typedef std::vector< OffsetType >	ComponentVectorType
typedef std::vector< OffsetType >	NeighborIndexContainer
typedef std::vector < NeighborIndexContainer >	NeighborIndexContainerContainer
Protected Member Functions
void	AnalyzeKernel ()
	BinaryMorphologyImageFilter ()
NeighborIndexContainer &	GetDifferenceSet (unsigned int code)
ComponentVectorConstIterator	KernelCCVectorBegin ()
ComponentVectorConstIterator	KernelCCVectorEnd ()
void	PrintSelf (std::ostream &os, Indent indent) const
virtual	~BinaryMorphologyImageFilter ()
Protected Attributes
bool	m_BoundaryToForeground
Private Member Functions
	BinaryMorphologyImageFilter (const Self &)
void	operator= (const Self &)
Private Attributes
OutputPixelType	m_BackgroundValue
InputPixelType	m_ForegroundValue
std::vector< OffsetType >	m_KernelCCVector
NeighborIndexContainerContainer	m_KernelDifferenceSets

Detailed Description

template<class TInputImage, class TOutputImage, class TKernel>
class itk::BinaryMorphologyImageFilter< TInputImage, TOutputImage, TKernel >

Base class for fast binary dilation and erosion.

BinaryMorphologyImageFilter is a base class for fast binary morphological operations. The implementation of this class and its subclasses are based on the papers:

L.Vincent "Morphological transformations of binary images with arbitrary structuring elements", and

N.Nikopoulos et al. "An efficient algorithm for 3d binary morphological transformations with 3d structuring elements for arbitrary size and shape". IEEE Transactions on Image Processing. Vol. 9. No. 3. 2000. pp. 283-286.

Gray scale images can be processed as binary images by selecting a "ForegroundValued" (which subclasses may alias as "DilateValue" or "ErodeValue"). Pixel not matching the foreground value are considered "background". This is useful in processing segmented images where all pixels in segment #1 have value 1 and pixels in segment #2 have value 2, etc. A particular "segment number" can be processed. ForegroundValue defaults to the maximum possible value of the PixelType.

The structuring element is assumed to be composed of binary values (zero or one). Only elements of the structuring element having values > 0 are candidates for affecting the center pixel. A reasonable choice of structuring element is itk::BinaryBallStructuringElement.

Description of the algorithm: ---------------------------------------------- Let's consider the set of the ON elements of the input image as X.

Let's consider the structuring element as B = {B0, B1, ..., Bn}, where Bi denotes a connected component of B.

Let's consider bi, i in [0,n], an arbitrary point of Bi.

We use hence the next property in order to compute minkoswki addition ( which will be written (+) ):

X (+) B = ( Xb0 UNION Xb1 UNION ... Xbn ) UNION ( BORDER(X) (+) B ),

where Xbi is the set X translated with respect to vector bi :

Xbi ={ x + bi, x belongs to X }

where BORDER(X) is the extracted border of X ( 8 connectivity in 2D, 26 in 3D )

Our implementation for dilation is defined as:

X (+) SYM(B) = DILATION(X)_B

Where DILATION(X)_B is the dilation of set with structuring element B. Where SYM(B) is the symmetric of the structuring element relatively to its center.

This code was contributed by Jerome Schmid from the University of Strasbourg who provided a fast dilation implementation. Gaetan Lehmann from INRA de Jouy-en-Josas then provided a fast erosion implementaton based on Jerome's implementation. The common portions of these two implementations were then placed in this superclass.

See also:: ImageToImageFilter BinaryErodeImageFilter BinaryDilateImageFilter

Definition at line 103 of file itkBinaryMorphologyImageFilter.h.