#include <otbMaximumAutocorrelationFactorImageFilter.h>

Inheritance diagram for otb::MaximumAutocorrelationFactorImageFilter< TInputImage, TOutputImage >:

Collaboration diagram for otb::MaximumAutocorrelationFactorImageFilter< TInputImage, TOutputImage >:

Public Types
typedef itk::SmartPointer< const Self >	ConstPointer

typedef CovarianceEstimatorType::Pointer	CovarianceEstimatorPointer

typedef StreamingStatisticsVectorImageFilter< InternalImageType >	CovarianceEstimatorType

typedef InputImageType::ConstPointer	InputImageConstPointer

typedef InputImageRegionType::IndexType	InputImageIndexType

typedef InputImageType::PixelType	InputImagePixelType

typedef InputImageType::Pointer	InputImagePointer

typedef InputImageType::RegionType	InputImageRegionType

typedef InputImageRegionType::SizeType	InputImageSizeType

typedef TInputImage	InputImageType

typedef InputImageType::ValueType	InputImageValueType

typedef InputImageType::InternalPixelType	InputInternalPixelType

typedef VectorImage< InternalPixelType, 2 >	InternalImageType

typedef MatrixType::InternalMatrixType	InternalMatrixType

typedef itk::NumericTraits< InputInternalPixelType >::RealType	InternalPixelType

typedef CovarianceEstimatorType::MatrixObjectType	MatrixObjectType

typedef MatrixObjectType::ComponentType	MatrixType

typedef OutputImageType::PixelType	OutputImagePixelType

typedef OutputImageType::Pointer	OutputImagePointer

typedef OutputImageType::RegionType	OutputImageRegionType

typedef TOutputImage	OutputImageType

typedef itk::SmartPointer< Self >	Pointer

typedef VectorType::ValueType	RealType

typedef MaximumAutocorrelationFactorImageFilter	Self

typedef itk::ImageToImageFilter< TInputImage, TOutputImage >	Superclass

typedef CovarianceEstimatorType::RealPixelType	VectorType

typedef vnl_matrix< RealType >	VnlMatrixType

typedef vnl_vector< RealType >	VnlVectorType

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

virtual VnlVectorType	GetAutoCorrelation ()

virtual const char *	GetNameOfClass () const

virtual VnlMatrixType	GetV ()

	itkGetObjectMacro (CovarianceEstimator, CovarianceEstimatorType)

	itkGetObjectMacro (CovarianceEstimatorH, CovarianceEstimatorType)

	itkGetObjectMacro (CovarianceEstimatorV, CovarianceEstimatorType)

Static Public Member Functions
static Pointer	New ()

Protected Member Functions
void	GenerateOutputInformation () override

	MaximumAutocorrelationFactorImageFilter ()

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

	~MaximumAutocorrelationFactorImageFilter () override

Private Member Functions
	MaximumAutocorrelationFactorImageFilter (const Self &)=delete

void	operator= (const Self &)=delete

Private Attributes
VnlVectorType	m_AutoCorrelation

CovarianceEstimatorPointer	m_CovarianceEstimator

CovarianceEstimatorPointer	m_CovarianceEstimatorH

CovarianceEstimatorPointer	m_CovarianceEstimatorV

VnlVectorType	m_Mean

VnlMatrixType	m_V

Detailed Description

template<class TInputImage, class TOutputImage>
class otb::MaximumAutocorrelationFactorImageFilter< TInputImage, TOutputImage >

This filter implements the Maximum Autocorrelation Factor.

This filter implements the Maximum Autocorrelation Factor, based on the following work:

A. A. Nielsen, "Kernel maximum autocorrelation factor and minimum noise fraction transformations," IEEE Transactions on Image Processing, vol. 20, no. 3, pp. 612-624, (2011)

Maximum Autocorrelation Factor can be considered as a spatial extension of the PCA, in which new variates try to maximize auto-correlation between neighboring pixels instead of variance. Though the inverse transform can be computed, this filter only provides the forward transform for now.

The GetV() method allows retrieving the linear combinations used to generate new variates, and the GetAutoCorrelation() method allows retrieving the auto-correlation associated to each variate.

This filter has been implemented from the Matlab code kindly made available by the authors here: http://www2.imm.dtu.dk/~aa/software.html

this filter have been validated by comparing the output image to the output produced by the Matlab code, and the reference image for testing has been generated from the Matlab code using Octave.

,

See also: otbPCAImageFilter

Definition at line 69 of file otbMaximumAutocorrelationFactorImageFilter.h.