This filter performs pixelwise voting among an arbitrary number of input images, where each of them represents a segmentation of the same scene (i.e., image). More...

#include <iAParametrizableLabelVotingImageFilter.h>

Inheritance diagram for iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >:

[legend]

Collaboration diagram for iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >:

[legend]

Public Types
typedef itk::ImageRegionConstIterator< DoubleImg >	ConstDblIt

typedef itk::SmartPointer< const Self >	ConstPointer

typedef itk::Image< double, 3 >	DoubleImg

typedef InputImageType::ConstPointer	InputImagePointer

typedef TInputImage	InputImageType
	Typedefs.

typedef TInputImage::PixelType	InputPixelType

typedef unsigned long	LabelCountType

typedef OutputImageType::Pointer	OutputImagePointer

typedef Superclass::OutputImageRegionType	OutputImageRegionType

typedef TOutputImage	OutputImageType

typedef TOutputImage::PixelType	OutputPixelType
	Extract some information from the image types.

typedef itk::SmartPointer< Self >	Pointer

typedef iAParametrizableLabelVotingImageFilter	Self
	Standard class typedefs.

typedef itk::ImageToImageFilter< TInputImage, TOutputImage >	Superclass

Public Member Functions
OutputPixelType	GetLabelForUndecidedPixels () const
	Get label value used for undecided pixels.

DoubleImg::Pointer	GetNumbers (int mode)

double	GetUndecided () const

	itkNewMacro (Self)
	Method for creation through the object factory.

	itkStaticConstMacro (ImageDimension, int, TOutputImage::ImageDimension)

	itkStaticConstMacro (InputImageDimension, int, TInputImage::ImageDimension)
	Extract some information from the image types.

	itkTypeMacro (iAParametrizableLabelVotingImageFilter, ImageToImageFilter)
	Run-time type information (and related methods)

void	SetAbsoluteMinimumPercentage (double p)
	Sets the percentage that has to be achieved as a minimum for the majority vote to be accepted as majority.

void	SetInputLabelVotersSet (std::set< std::pair< int, int > > inputLabelVotersSet)
	sets the data determining which input is considered for voting on which label At each pixel the label value is checked, and an input is only considered if this input is considered a suitable voter for this label (i.e., if the pair <label, input index> is contained in the set given here)

void	SetInputLabelWeightMap (std::map< std::pair< int, int >, double > inputLabelWeightMap)
	set a weight for a pair of <label, input index> only used if WeightType LabelBased is used (see WeightType, SetWeightType)

void	SetLabelForUndecidedPixels (const OutputPixelType l)
	Set label value for undecided pixels.

void	SetMaxPixelEntropy (double e)
	set a maximum threshold for the pixel entropy

void	SetMinimumDifferencePercentage (double p)
	set a minimum threshold for the percentage difference between first and second best guess

void	SetMinimumRatio (double r)
	set a minimum threshold on the ratio between first and second best guess

void	SetProbabilityImages (size_t inputIdx, std::vector< DoubleImg::Pointer > const &probImgs)
	set probabilities for each label for one input in order for probability-related functionality to work, this method needs to be called once with each input idx

void	SetWeightType (WeightType weightType)
	determine the type of weight given each input in the voting process see WeightType for more details

void	UnsetLabelForUndecidedPixels ()
	Unset label value for undecided pixels and turn on automatic selection.

Protected Member Functions
	iAParametrizableLabelVotingImageFilter ()

virtual	~iAParametrizableLabelVotingImageFilter ()

void	BeforeThreadedGenerateData () override
	Determine maximum label value in all input images and initialize global data.

InputPixelType	ComputeMaximumInputValue ()
	Determine maximum value among all input images' pixels.

void	PrintSelf (std::ostream &, itk::Indent) const override

void	ThreadedGenerateData (const OutputImageRegionType &outputRegionForThread, itk::ThreadIdType threadId) override

Detailed Description

template<typename TInputImage, typename TOutputImage = TInputImage>
class iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >

This filter performs pixelwise voting among an arbitrary number of input images, where each of them represents a segmentation of the same scene (i.e., image).

Label voting is a simple method of classifier combination applied to image segmentation. Typically, the accuracy of the combined segmentation exceeds the accuracy of any of the input segmentations. Voting is therefore commonly used as a way of boosting segmentation performance.

The use of label voting for combination of multiple segmentations is described in

T. Rohlfing and C. R. Maurer, Jr., "Multi-classifier framework for atlas-based image segmentation," Pattern Recognition Letters, 2005.

INPUTS: All input volumes to this filter must be segmentations of an image, that is, they must have discrete pixel values where each value represents a different segmented object.

Input volumes must all contain the same size RequestedRegions. Not all input images must contain all possible labels, but all label values must have the same meaning in all images.

OUTPUTS: The voting filter produces a single output volume. Each output pixel contains the label that occurred most often among the labels assigned to this pixel in all the input volumes, that is, the label that received the maximum number of "votes" from the input pixels.. If the maximum number of votes is not unique, i.e., if more than one label have a maximum number of votes, an "undecided" label is assigned to that output pixel.

By default, the label used for undecided pixels is the maximum label value used in the input images plus one. Since it is possible for an image with 8 bit pixel values to use all 256 possible label values, it is permissible to combine 8 bit (i.e., byte) images into a 16 bit (i.e., short) output image.

PARAMETERS: The label used for "undecided" labels can be set using SetLabelForUndecidedPixels. This functionality can be unset by calling UnsetLabelForUndecidedPixels.

Author: Torsten Rohlfing, SRI International, Neuroscience Program

Member Typedef Documentation

◆ ConstDblIt

template<typename TInputImage , typename TOutputImage = TInputImage>

itk::ImageRegionConstIterator<DoubleImg> iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::ConstDblIt

◆ ConstPointer

template<typename TInputImage , typename TOutputImage = TInputImage>

itk::SmartPointer< const Self > iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::ConstPointer

◆ DoubleImg

template<typename TInputImage , typename TOutputImage = TInputImage>

itk::Image<double, 3> iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::DoubleImg

◆ InputImagePointer

template<typename TInputImage , typename TOutputImage = TInputImage>

InputImageType::ConstPointer iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::InputImagePointer

◆ InputImageType

template<typename TInputImage , typename TOutputImage = TInputImage>

TInputImage iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::InputImageType

Typedefs.

◆ InputPixelType

template<typename TInputImage , typename TOutputImage = TInputImage>

TInputImage::PixelType iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::InputPixelType

◆ LabelCountType

template<typename TInputImage , typename TOutputImage = TInputImage>

unsigned long iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::LabelCountType

◆ OutputImagePointer

template<typename TInputImage , typename TOutputImage = TInputImage>

OutputImageType::Pointer iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::OutputImagePointer

◆ OutputImageRegionType

template<typename TInputImage , typename TOutputImage = TInputImage>

Superclass::OutputImageRegionType iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::OutputImageRegionType

◆ OutputImageType

template<typename TInputImage , typename TOutputImage = TInputImage>

TOutputImage iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::OutputImageType

◆ OutputPixelType

template<typename TInputImage , typename TOutputImage = TInputImage>

TOutputImage::PixelType iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::OutputPixelType

Extract some information from the image types.

Dimensionality of the two images is assumed to be the same.

◆ Pointer

template<typename TInputImage , typename TOutputImage = TInputImage>

itk::SmartPointer< Self > iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::Pointer

◆ Self

template<typename TInputImage , typename TOutputImage = TInputImage>

iAParametrizableLabelVotingImageFilter iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::Self

Standard class typedefs.

◆ Superclass

template<typename TInputImage , typename TOutputImage = TInputImage>

itk::ImageToImageFilter< TInputImage, TOutputImage > iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::Superclass

Constructor & Destructor Documentation

◆ iAParametrizableLabelVotingImageFilter()

template<typename TInputImage , typename TOutputImage >

iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::iAParametrizableLabelVotingImageFilter ( )

protected

◆ ~iAParametrizableLabelVotingImageFilter()

template<typename TInputImage , typename TOutputImage = TInputImage>

virtual iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::~iAParametrizableLabelVotingImageFilter ( )

inlineprotectedvirtual

Member Function Documentation

◆ BeforeThreadedGenerateData()

template<typename TInputImage , typename TOutputImage >

void iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::BeforeThreadedGenerateData ( )

overrideprotected

Determine maximum label value in all input images and initialize global data.

◆ ComputeMaximumInputValue()

template<typename TInputImage , typename TOutputImage >

iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::InputPixelType iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::ComputeMaximumInputValue ( )

protected

Determine maximum value among all input images' pixels.

◆ GetLabelForUndecidedPixels()

template<typename TInputImage , typename TOutputImage = TInputImage>

OutputPixelType iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::GetLabelForUndecidedPixels ( ) const

inline

Get label value used for undecided pixels.

After updating the filter, this function returns the actual label value used for undecided pixels in the current output. Note that this value is overwritten when SetLabelForUndecidedPixels is called and the new value only becomes effective upon the next filter update.

◆ GetNumbers()

template<typename TInputImage , typename TOutputImage = TInputImage>

DoubleImg::Pointer iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::GetNumbers ( int mode )

inline

◆ GetUndecided()

template<typename TInputImage , typename TOutputImage = TInputImage>

double iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::GetUndecided ( ) const

inline

◆ itkNewMacro()

template<typename TInputImage , typename TOutputImage = TInputImage>

iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::itkNewMacro ( Self )

Method for creation through the object factory.

◆ itkStaticConstMacro() [1/2]

template<typename TInputImage , typename TOutputImage = TInputImage>

iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::itkStaticConstMacro	(	ImageDimension	,
		int	,
		TOutputImage::ImageDimension	)

◆ itkStaticConstMacro() [2/2]

template<typename TInputImage , typename TOutputImage = TInputImage>

iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::itkStaticConstMacro	(	InputImageDimension	,
		int	,
		TInputImage::ImageDimension	)

Extract some information from the image types.

Dimensionality of the two images is assumed to be the same.

◆ itkTypeMacro()

template<typename TInputImage , typename TOutputImage = TInputImage>

iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::itkTypeMacro	(	iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >	,
		ImageToImageFilter	)

Run-time type information (and related methods)

◆ PrintSelf()

template<typename TInputImage , typename TOutputImage >

void iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::PrintSelf	(	std::ostream &	os,
		itk::Indent	indent ) const

overrideprotected

◆ SetAbsoluteMinimumPercentage()

template<typename TInputImage , typename TOutputImage = TInputImage>

void iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::SetAbsoluteMinimumPercentage ( double p )

inline

Sets the percentage that has to be achieved as a minimum for the majority vote to be accepted as majority.

Parameters

p	the percentage (in the interval 0..1); 0 -> no minimum percentage 1 -> all images have to agree on a label

◆ SetInputLabelVotersSet()

template<typename TInputImage , typename TOutputImage = TInputImage>

void iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::SetInputLabelVotersSet ( std::set< std::pair< int, int > > inputLabelVotersSet )

inline

sets the data determining which input is considered for voting on which label At each pixel the label value is checked, and an input is only considered if this input is considered a suitable voter for this label (i.e., if the pair <label, input index> is contained in the set given here)

◆ SetInputLabelWeightMap()

template<typename TInputImage , typename TOutputImage = TInputImage>

void iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::SetInputLabelWeightMap ( std::map< std::pair< int, int >, double > inputLabelWeightMap )

inline

set a weight for a pair of <label, input index> only used if WeightType LabelBased is used (see WeightType, SetWeightType)

◆ SetLabelForUndecidedPixels()

template<typename TInputImage , typename TOutputImage = TInputImage>

void iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::SetLabelForUndecidedPixels ( const OutputPixelType l )

inline

Set label value for undecided pixels.

◆ SetMaxPixelEntropy()

template<typename TInputImage , typename TOutputImage = TInputImage>

void iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::SetMaxPixelEntropy ( double e )

inline

set a maximum threshold for the pixel entropy

◆ SetMinimumDifferencePercentage()

template<typename TInputImage , typename TOutputImage = TInputImage>

void iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::SetMinimumDifferencePercentage ( double p )

inline

set a minimum threshold for the percentage difference between first and second best guess

◆ SetMinimumRatio()

template<typename TInputImage , typename TOutputImage = TInputImage>

void iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::SetMinimumRatio ( double r )

inline

set a minimum threshold on the ratio between first and second best guess

◆ SetProbabilityImages()

template<typename TInputImage , typename TOutputImage = TInputImage>

void iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::SetProbabilityImages	(	size_t	inputIdx,
		std::vector< DoubleImg::Pointer > const &	probImgs )

inline

set probabilities for each label for one input in order for probability-related functionality to work, this method needs to be called once with each input idx

◆ SetWeightType()

template<typename TInputImage , typename TOutputImage = TInputImage>

void iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::SetWeightType ( WeightType weightType )

inline

determine the type of weight given each input in the voting process see WeightType for more details

◆ ThreadedGenerateData()

template<typename TInputImage , typename TOutputImage >

void iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::ThreadedGenerateData	(	const OutputImageRegionType &	outputRegionForThread,
		itk::ThreadIdType	threadId )

overrideprotected

◆ UnsetLabelForUndecidedPixels()

template<typename TInputImage , typename TOutputImage = TInputImage>

void iAParametrizableLabelVotingImageFilter< TInputImage, TOutputImage >::UnsetLabelForUndecidedPixels ( )

inline

Unset label value for undecided pixels and turn on automatic selection.

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

Toolkit/LabelVoting/iAParametrizableLabelVotingImageFilter.h
Toolkit/LabelVoting/iAParametrizableLabelVotingImageFilter.hxx

Public Types

Public Member Functions

Protected Member Functions

Detailed Description

Member Typedef Documentation

◆ ConstDblIt

◆ ConstPointer

◆ DoubleImg

◆ InputImagePointer

◆ InputImageType

◆ InputPixelType

◆ LabelCountType

◆ OutputImagePointer

◆ OutputImageRegionType

◆ OutputImageType

◆ OutputPixelType

◆ Pointer

◆ Self

◆ Superclass

Constructor & Destructor Documentation

◆ iAParametrizableLabelVotingImageFilter()

◆ ~iAParametrizableLabelVotingImageFilter()

Member Function Documentation

◆ BeforeThreadedGenerateData()

◆ ComputeMaximumInputValue()

◆ GetLabelForUndecidedPixels()

◆ GetNumbers()

◆ GetUndecided()

◆ itkNewMacro()

◆ itkStaticConstMacro() [1/2]

◆ itkStaticConstMacro() [2/2]

◆ itkTypeMacro()

◆ PrintSelf()

◆ SetAbsoluteMinimumPercentage()

◆ SetInputLabelVotersSet()

◆ SetInputLabelWeightMap()

◆ SetLabelForUndecidedPixels()

◆ SetMaxPixelEntropy()

◆ SetMinimumDifferencePercentage()

◆ SetMinimumRatio()

◆ SetProbabilityImages()

◆ SetWeightType()

◆ ThreadedGenerateData()

◆ UnsetLabelForUndecidedPixels()