This class computes texture feature coefficients from a grey level co-occurrence matrix. More...

#include <itkEnhancedHistogramToTextureFeaturesFilter.h>

Inheritance diagram for itk::Statistics::EnhancedHistogramToTextureFeaturesFilter< THistogram >:

Collaboration diagram for itk::Statistics::EnhancedHistogramToTextureFeaturesFilter< THistogram >:

Public Types
enum	TextureFeatureName { Energy, Entropy, Correlation, InverseDifferenceMoment, Inertia, ClusterShade, ClusterProminence, HaralickCorrelation, Autocorrelation, Contrast, Dissimilarity, MaximumProbability, InverseVariance, Homogeneity1, ClusterTendency, Variance, SumAverage, SumEntropy, SumVariance, DifferenceAverage, DifferenceEntropy, DifferenceVariance, InverseDifferenceMomentNormalized, InverseDifferenceNormalized, InverseDifference, InvalidFeatureName }

typedef EnhancedHistogramToTextureFeaturesFilter	Self

typedef ProcessObject	Superclass

typedef SmartPointer< Self >	Pointer

typedef SmartPointer< const Self >	ConstPointer

typedef THistogram	HistogramType

typedef HistogramType::Pointer	HistogramPointer

typedef HistogramType::ConstPointer	HistogramConstPointer

typedef HistogramType::MeasurementType	MeasurementType

typedef HistogramType::MeasurementVectorType	MeasurementVectorType

typedef HistogramType::IndexType	IndexType

typedef HistogramType::AbsoluteFrequencyType	AbsoluteFrequencyType

typedef HistogramType::RelativeFrequencyType	RelativeFrequencyType

typedef HistogramType::TotalAbsoluteFrequencyType	TotalAbsoluteFrequencyType

typedef HistogramType::TotalRelativeFrequencyType	TotalRelativeFrequencyType

typedef std::vector< RelativeFrequencyType >	RelativeFrequencyContainerType

typedef DataObject::Pointer	DataObjectPointer

typedef SimpleDataObjectDecorator< MeasurementType >	MeasurementObjectType

Public Member Functions
virtual const char *	GetClassName () const

void	SetInput (const HistogramType *histogram)

const HistogramType *	GetInput () const

MeasurementType	GetEnergy () const

const MeasurementObjectType *	GetEnergyOutput () const

MeasurementType	GetEntropy () const

const MeasurementObjectType *	GetEntropyOutput () const

MeasurementType	GetCorrelation () const

const MeasurementObjectType *	GetCorrelationOutput () const

MeasurementType	GetInverseDifferenceMoment () const

const MeasurementObjectType *	GetInverseDifferenceMomentOutput () const

MeasurementType	GetInertia () const

const MeasurementObjectType *	GetInertiaOutput () const

MeasurementType	GetClusterShade () const

const MeasurementObjectType *	GetClusterShadeOutput () const

MeasurementType	GetClusterProminence () const

const MeasurementObjectType *	GetClusterProminenceOutput () const

MeasurementType	GetHaralickCorrelation () const

const MeasurementObjectType *	GetHaralickCorrelationOutput () const

const MeasurementObjectType *	GetAutocorrelationOutput () const

MeasurementType	GetAutocorrelation () const

const MeasurementObjectType *	GetContrastOutput () const

MeasurementType	GetContrast () const

const MeasurementObjectType *	GetDissimilarityOutput () const

MeasurementType	GetDissimilarity () const

const MeasurementObjectType *	GetMaximumProbabilityOutput () const

MeasurementType	GetMaximumProbability () const

const MeasurementObjectType *	GetInverseVarianceOutput () const

MeasurementType	GetInverseVariance () const

const MeasurementObjectType *	GetHomogeneity1Output () const

MeasurementType	GetHomogeneity1 () const

const MeasurementObjectType *	GetClusterTendencyOutput () const

MeasurementType	GetClusterTendency () const

const MeasurementObjectType *	GetVarianceOutput () const

MeasurementType	GetVariance () const

const MeasurementObjectType *	GetSumAverageOutput () const

MeasurementType	GetSumAverage () const

const MeasurementObjectType *	GetSumEntropyOutput () const

MeasurementType	GetSumEntropy () const

const MeasurementObjectType *	GetSumVarianceOutput () const

MeasurementType	GetSumVariance () const

const MeasurementObjectType *	GetDifferenceAverageOutput () const

MeasurementType	GetDifferenceAverage () const

const MeasurementObjectType *	GetDifferenceEntropyOutput () const

MeasurementType	GetDifferenceEntropy () const

const MeasurementObjectType *	GetDifferenceVarianceOutput () const

MeasurementType	GetDifferenceVariance () const

const MeasurementObjectType *	GetInverseDifferenceMomentNormalizedOutput () const

MeasurementType	GetInverseDifferenceMomentNormalized () const

const MeasurementObjectType *	GetInverseDifferenceNormalizedOutput () const

MeasurementType	GetInverseDifferenceNormalized () const

const MeasurementObjectType *	GetInverseDifferenceOutput () const

MeasurementType	GetInverseDifference () const

MeasurementType	GetFeature (TextureFeatureName name)

Static Public Member Functions
static Pointer	New ()

Protected Types
typedef ProcessObject::DataObjectPointerArraySizeType	DataObjectPointerArraySizeType

Protected Member Functions
	EnhancedHistogramToTextureFeaturesFilter ()

	~EnhancedHistogramToTextureFeaturesFilter ()

virtual void	PrintSelf (std::ostream &os, Indent indent) const ITK_OVERRIDE

virtual DataObjectPointer	MakeOutput (DataObjectPointerArraySizeType) ITK_OVERRIDE

virtual void	GenerateData () ITK_OVERRIDE

Detailed Description

template<typename THistogram>
class itk::Statistics::EnhancedHistogramToTextureFeaturesFilter< THistogram >

This class computes texture feature coefficients from a grey level co-occurrence matrix.

This class computes features that summarize image texture, given a grey level co-occurrence matrix (generated by a ScalarImageToCooccurrenceMatrixFilter or related class).

The features calculated are as follows (where \( g(i, j) \) is the element in cell i, j of a a normalized GLCM):

"Energy" \( = f_1 = \sum_{i,j}g(i, j)^2 \)

"Entropy" \( = f_2 = -\sum_{i,j}g(i, j) \log_2 g(i, j)\), or 0 if \(g(i, j) = 0\)

"Correlation" \( = f_3 = \sum_{i,j}\frac{(i - \mu)(j - \mu)g(i, j)}{\sigma^2} \)

"Difference Moment" \(= f_4 = \sum_{i,j}\frac{1}{1 + (i - j)^2}g(i, j) \)

"Inertia" \( = f_5 = \sum_{i,j}(i - j)^2g(i, j) \) (sometimes called "contrast.")

"Cluster Shade" \( = f_6 = \sum_{i,j}((i - \mu) + (j - \mu))^3 g(i, j) \)

"Cluster Prominence" \( = f_7 = \sum_{i,j}((i - \mu) + (j - \mu))^4 g(i, j) \)

"Haralick's Correlation" \( = f_8 = \frac{\sum_{i,j}(i, j) g(i, j) -\mu_t^2}{\sigma_t^2} \) where \(\mu_t\) and \(\sigma_t\) are the mean and standard deviation of the row (or column, due to symmetry) sums.

Above, \( \mu = \) (weighted pixel average) \( = \sum_{i,j}i \cdot g(i, j) = \sum_{i,j}j \cdot g(i, j) \) (due to matrix summetry), and

\( \sigma = \) (weighted pixel variance) \( = \sum_{i,j}(i - \mu)^2 \cdot g(i, j) = \sum_{i,j}(j - \mu)^2 \cdot g(i, j) \) (due to matrix summetry)

A good texture feature set to use is the Conners, Trivedi and Harlow set: features 1, 2, 4, 5, 6, and 7. There is some correlation between the various features, so using all of them at the same time is not necessarialy a good idea.

NOTA BENE: The input histogram will be forcably normalized! This algorithm takes three passes through the input histogram if the histogram was already normalized, and four if not.

Web references:

http://www.cssip.uq.edu.au/meastex/www/algs/algs/algs.html http://www.ucalgary.ca/~mhallbey/texture/texture_tutorial.html

Print references:

Haralick, R.M., K. Shanmugam and I. Dinstein. 1973. Textural Features for Image Classification. IEEE Transactions on Systems, Man and Cybernetics. SMC-3(6):610-620.

Haralick, R.M. 1979. Statistical and Structural Approaches to Texture. Proceedings of the IEEE, 67:786-804.

R.W. Conners and C.A. Harlow. A Theoretical Comaprison of Texture Algorithms. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2:204-222, 1980.

R.W. Conners, M.M. Trivedi, and C.A. Harlow. Segmentation of a High-Resolution Urban Scene using Texture Operators. Computer Vision, Graphics and Image Processing, 25:273-310, 1984.

See also: ScalarImageToCooccurrenceMatrixFilter; ScalarImageToTextureFeaturesFilter

Author: Zachary Pincus

Definition at line 108 of file itkEnhancedHistogramToTextureFeaturesFilter.h.