#include <otbDSFusionOfClassifiersImageFilter.h>

Inheritance diagram for otb::DSFusionOfClassifiersImageFilter< TInputImage, TOutputImage, TMaskImage >:

Collaboration diagram for otb::DSFusionOfClassifiersImageFilter< TInputImage, TOutputImage, TMaskImage >:

Public Types
typedef std::map< LabelType, unsigned int >	ClassifierHistogramType

typedef itk::SmartPointer< const Self >	ConstPointer

typedef InputImageType::ConstPointer	InputImageConstPointerType

typedef TInputImage	InputImageType

typedef InputImageType::InternalPixelType	InternalPixelType

typedef std::map< LabelType, MassType >	LabelMassMapType

typedef OutputImageType::PixelType	LabelType

typedef MaskImageType::ConstPointer	MaskImageConstPointerType

typedef MaskImageType::Pointer	MaskImagePointerType

typedef TMaskImage	MaskImageType

typedef double	MassType

typedef OutputImageType::Pointer	OutputImagePointerType

typedef OutputImageType::RegionType	OutputImageRegionType

typedef TOutputImage	OutputImageType

typedef InputImageType::PixelType	PixelType

typedef itk::SmartPointer< Self >	Pointer

typedef DSFusionOfClassifiersImageFilter	Self

typedef itk::ImageToImageFilter< TInputImage, TOutputImage >	Superclass

typedef InputImageType::InternalPixelType	ValueType

typedef std::vector< LabelMassMapType >	VectorOfMapOfMassesOfBeliefType

Public Member Functions
virtual ::itk::LightObject::Pointer	CreateAnother (void) const

virtual const char *	GetNameOfClass () const

Static Public Member Functions
static Pointer	New ()

unsigned int	m_NumberOfClassifiers

ClassifierHistogramType	m_Universe

unsigned int	m_NumberOfClassesInUniverse

VectorOfMapOfMassesOfBeliefType	m_VectorOfMapMOBs

std::vector< MassType >	m_VectorOfUniverseMOBs

LabelType	m_LabelForNoDataPixels

LabelType	m_LabelForUndecidedPixels

virtual void	SetLabelForNoDataPixels (LabelType _arg)

virtual LabelType	GetLabelForNoDataPixels ()

virtual void	SetLabelForUndecidedPixels (LabelType _arg)

virtual LabelType	GetLabelForUndecidedPixels ()

void	SetInputMask (const MaskImageType *mask)

void	SetInputMapsOfMassesOfBelief (const VectorOfMapOfMassesOfBeliefType *ptrVectorOfMapOfMassesOfBelief)

const MaskImageType *	GetInputMask (void)

const VectorOfMapOfMassesOfBeliefType *	GetInputMapsOfMassesOfBelief (void)

const LabelType	OptimizedDSMassCombination (PixelType vectorPixelValue)

	DSFusionOfClassifiersImageFilter ()

	~DSFusionOfClassifiersImageFilter () override

void	GenerateOutputInformation () override

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

void	BeforeThreadedGenerateData () override

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

	DSFusionOfClassifiersImageFilter (const Self &)=delete

void	operator= (const Self &)=delete

Detailed Description

template<class TInputImage, class TOutputImage, class TMaskImage = TOutputImage>
class otb::DSFusionOfClassifiersImageFilter< TInputImage, TOutputImage, TMaskImage >

Performs a fusion of n classification maps using the Dempster Shafer (noted DS) rule of combination.

This filter performs a fusion of n classification maps using the Dempster Shafer rule of combination from an otbVectorImage with n channels, each one corresponding to an input classification map to be fused. This otbVectorImage can be obtained with the help of a preliminary use of the otbImageListToVectorImageFilter.

Moreover, the otbDSFusionOfClassifiersImageFilter needs as additional input a std::vector containing the n std::maps of Masses Of Belief (noted MOB). Each std::map<Label, MOB> in this input std::vector is related to one classifier and links each label to its mass of belief in this classifier.

The output label image resulting from the DS fusion of classification maps only has one channel (monoband image).

The recursive optimized Dempster Shafer combination of masses of belief is performed in 2 steps for each channel component of each pixel of the input otbVectorImage within the input mask, and is based on:

[1] L. Xu, A. Krzyzak, and C.Y. Suen, "Methods of combining multiple classifiers and their applications to handwriting recognition," Systems, Man and Cybernetics, IEEE Transactions on vol. 22, no.3, pp. 418-435, May/Jun 1992.

[2] D. Ruta and B. Gabrys, "An Overview of Classifier Fusion Methods," Computing and Information Systems, vol. 7 pp.1-10, University of Paisley, 2000.

[3] C.R. Parikh, M.J. Pont, and N.B. Jones, "Application of Dempster-Shafer theory in condition monitoring systems: A case study," Pattern Recognition Letters, vol. 22 (6-7): pp. 777-785, 2001.

See also: MassOfBelief

Definition at line 69 of file otbDSFusionOfClassifiersImageFilter.h.