de.unihalle.informatik.MiToBo.features.texture

Class FeatureCalculatorHaralickMeasures

java.lang.Object

de.unihalle.informatik.Alida.operator.ALDOperator

de.unihalle.informatik.MiToBo.core.operator.MTBOperator

de.unihalle.informatik.MiToBo.features.FeatureCalculator

de.unihalle.informatik.MiToBo.features.texture.FeatureCalculatorHaralickMeasures

All Implemented Interfaces:

de.unihalle.informatik.Alida.datatypes.ALDConfigurationValidator, de.unihalle.informatik.Alida.operator.events.ALDOperatorExecutionProgressEventListener, EventListener

@ALDAOperator(genericExecutionMode=ALL,
              level=STANDARD,
              allowBatchMode=false,
              shortDescription="Calculates set of Haralick co-occurrence texture features from an image.")
 @ALDDerivedClass
public class FeatureCalculatorHaralickMeasures
extends FeatureCalculator

Calculates set of Haralick co-occurrence texture features from an image.

The operator expects that the input image does not contain more than 256 different gray-values, color images are completely rejected. For saving memory the operator first checks the maximal gray-value in the image and updates the size of the gray-level co-occurence matrices accordingly. Subsequently, the following measures are calculated:

• homogenity: $\sum_i \sum_j p(i,j)^2$
• contrast: $\sum_i \sum_j |i-j|^2 \cdot p(i,j)$
• local homogeneity: $\sum_i \sum_j \frac{1}{1 + (i-j)^2} p(i,j)$
• entropy: $- \sum_i \sum_j p(i,j) \cdot \log p(i,j)$
• correlation: $\sum_i \sum_j \frac{i \cdot j \cdot p(i,j) - \mu_x \cdot \mu_y}{\sigma_x \cdot \sigma_y}$
• auto-correlation: $\sum_i \sum_j i \cdot j \cdot p(i,j)$
• dissimilarity: $\sum_i \sum_j |i - j| \cdot p(i,j)$
• cluster shade: $\sum_i \sum_j (i + j - \mu_x - \mu_y)^3 \cdot p(i,j)$
• cluster prominence: $\sum_i \sum_j (i + j - \mu_x - \mu_y)^4 \cdot p(i,j)$
• maximum probability: $\max p(i,j)$

For details about these measures, refer to one of the following papers:

• R.M. Haralick, K. Shanmugan, and I. Dinstein, Textural Features for Image Classification, IEEE Trans. on Systems, Man and Cybernetics, vol. SMC-3, no. 6, pp. 610-621, 1973.
• F. Albregtsen, Statistical Texture Measures Computed from Gray Level Cooccurence Matrices, Technical Report, Image Processing Laboratory, Department of Informatics, University of Oslo, November 5, 2008.
• L.-K. Soh and C. Tsatsoulis, Texture Analysis of SAR Sea Ice Imagery Using Gray Level Co-Occurence Matrices, IEEE Trans. on Geoscience and Remote Sensing, vol. 37, no. 2, pp. 779-795, 1999.

Author:

Birgit Moeller, Elisabeth Piltz

Nested Class Summary

Nested Classes

Modifier and Type	Class and Description
static class	FeatureCalculatorHaralickMeasures.HaralickDirection Directions available for matrix calculations.

Nested classes/interfaces inherited from class de.unihalle.informatik.Alida.operator.ALDOperator

de.unihalle.informatik.Alida.operator.ALDOperator.HidingMode, de.unihalle.informatik.Alida.operator.ALDOperator.HistoryConstructionMode

Field Summary

Fields

Modifier and Type	Field and Description
private Vector<FeatureCalculatorHaralickMeasures.HaralickDirection>	directions Directions.
private int	distance Distance.
private boolean	isotropicCalcs Flag for isotropic calculations.
private boolean	thrinkMatrix Flag for thrinking matrix if possible.

Fields inherited from class de.unihalle.informatik.MiToBo.features.FeatureCalculator

inImg, resultObj

Fields inherited from class de.unihalle.informatik.Alida.operator.ALDOperator

completeDAG, name, operatorExecutionEventlistenerList, portHashAccess, verbose, versionProvider

Constructor Summary

Constructors

Constructor and Description
FeatureCalculatorHaralickMeasures() Default constructor.

Method Summary

Modifier and Type	Method and Description
protected double[]	calcFeatures(double[][] matrix) Extracts Haralick features from given matrix.
protected double[][]	calcMatrix(FeatureCalculatorHaralickMeasures.HaralickDirection dir) Extracts the co-occurence matrix for the given distance and direction.
void	doIsotropicCalcutations(boolean flag) Set flag for isotropic calculations.
String	getDocumentation()
protected FeatureCalculatorHaralickMeasuresResult	getResultDataObjectInvalid(int dim) Helper method to allow for requesting invalid dummy results.
void	operate()
void	setDirections(Vector<FeatureCalculatorHaralickMeasures.HaralickDirection> dirs) Directions to be analyzed.
void	setDistance(int dist) Specify distance to be used in calculations.
void	setFlagThrinkMatrix(boolean b) Flag for thrinking matrix if possible.
void	validateCustom()

Methods inherited from class de.unihalle.informatik.MiToBo.features.FeatureCalculator

getResultData, setInputImage

Methods inherited from class de.unihalle.informatik.MiToBo.core.operator.MTBOperator

readResolve

Methods inherited from class de.unihalle.informatik.Alida.operator.ALDOperator

addOperatorExecutionProgressEventListener, addParameter, addParameter, addParameterUnconditioned, fieldContained, fireOperatorExecutionProgressEvent, getALDPortHashAccessKey, getConstructionMode, getHidingMode, getInactiveParameterNames, getInInoutNames, getInInoutNames, getInNames, getInOutNames, getMissingRequiredInputs, getName, getNumParameters, getOutInoutNames, getOutNames, getParameter, getParameterDescriptor, getParameterDescriptorUnconditioned, getParameterNames, getParameterUnconditioned, getSupplementalNames, getVerbose, getVersion, handleOperatorExecutionProgressEvent, hasInOutParameters, hasParameter, isAnnotatedParameter, isConfigured, print, print, print, printInterface, printInterface, readHistory, reinitializeParameterDescriptors, removeOperatorExecutionProgressEventListener, removeParameter, runOp, runOp, runOp, setConstructionMode, setConstructionMode, setConstructionMode, setHidingMode, setName, setParameter, setParameterUnconditioned, setVerbose, toStringVerbose, unconfiguredItems, validate, validateGeneric, writeHistory, writeHistory, writeHistory

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

directions

@Parameter(label="Directions",
           required=true,
           direction=IN,
           dataIOOrder=0,
           mode=STANDARD,
           description="Desired directions.")
private Vector<FeatureCalculatorHaralickMeasures.HaralickDirection> directions

Directions.

Note that no checks are done if directions appear several times.

isotropicCalcs

@Parameter(label="Isotropic calculations",
           required=true,
           direction=IN,
           dataIOOrder=1,
           mode=STANDARD,
           description="Flag to enable isotropic calculations.")
private boolean isotropicCalcs

Flag for isotropic calculations.

If flag is true the average of all given directions is calculated.

distance

@Parameter(label="Distance",
           required=true,
           direction=IN,
           dataIOOrder=2,
           mode=STANDARD,
           description="Desired distance.")
private int distance

Distance.

thrinkMatrix

@Parameter(label="Reduce matrix size",
           required=true,
           direction=IN,
           dataIOOrder=4,
           mode=STANDARD,
           description="Thrinks the matrix if not the whole range of gray-values is actually used.")
private boolean thrinkMatrix

Flag for thrinking matrix if possible.

Constructor Detail

FeatureCalculatorHaralickMeasures

public FeatureCalculatorHaralickMeasures()
                                  throws de.unihalle.informatik.Alida.exceptions.ALDOperatorException

Default constructor.

Throws:

de.unihalle.informatik.Alida.exceptions.ALDOperatorException - Thrown in case of failure.

Method Detail

validateCustom

public void validateCustom()
                    throws de.unihalle.informatik.Alida.exceptions.ALDOperatorException

Overrides:

validateCustom in class de.unihalle.informatik.Alida.operator.ALDOperator

Throws:

de.unihalle.informatik.Alida.exceptions.ALDOperatorException

operate

public void operate()

Specified by:

operate in class de.unihalle.informatik.Alida.operator.ALDOperator

calcMatrix

protected double[][] calcMatrix(FeatureCalculatorHaralickMeasures.HaralickDirection dir)

Extracts the co-occurence matrix for the given distance and direction.

Parameters:

dir - Haralick direction to consider.

Returns:

Resulting (normalized) matrix.

calcFeatures

protected double[] calcFeatures(double[][] matrix)

Extracts Haralick features from given matrix.

Order of the features in the result vector is as follows: 0 = homogenity (also known as energy) 1 = contrast (|i-j|² * p(i,j)) 2 = local homogeneity 3 = entropy 4 = correlation 5 = auto-correlation 6 = dissimilarity 7 = cluster shade 8 = cluster prominence 9 = maximum probability

Parameters:

matrix - Normalized co-occurence matrix.

Returns:

Vector of features.

doIsotropicCalcutations

public void doIsotropicCalcutations(boolean flag)

Set flag for isotropic calculations.

Parameters:

flag - If true, isotropic calculations are done.

setDistance

public void setDistance(int dist)

Specify distance to be used in calculations.

Parameters:

dist - Distance to be used (in pixels).

setDirections

public void setDirections(Vector<FeatureCalculatorHaralickMeasures.HaralickDirection> dirs)

Directions to be analyzed.

Parameters:

dirs - Set of directions to consider.

See Also:

FeatureCalculatorHaralickMeasures.HaralickDirection

setFlagThrinkMatrix

public void setFlagThrinkMatrix(boolean b)

Flag for thrinking matrix if possible.

Parameters:

b - If true, matrix size is reduced, otherwise it remains 256x256.

getResultDataObjectInvalid

protected FeatureCalculatorHaralickMeasuresResult getResultDataObjectInvalid(int dim)

Description copied from class: FeatureCalculator

Helper method to allow for requesting invalid dummy results.

Specified by:

getResultDataObjectInvalid in class FeatureCalculator

Parameters:

dim - Desired dimension of output object.

Returns:

Invalid dummy result.

getDocumentation

public String getDocumentation()

Overrides:

getDocumentation in class de.unihalle.informatik.Alida.operator.ALDOperator