de.lmu.ifi.dbs.jfeaturelib.features

Class AbstractFeatureDescriptor

java.lang.Object

de.lmu.ifi.dbs.jfeaturelib.features.AbstractFeatureDescriptor

All Implemented Interfaces:

Descriptor, FeatureDescriptor

Direct Known Subclasses:

AdaptiveGridResolution, AutoColorCorrelogram, CEDD, CentroidBoundaryDistance, CentroidFeature, ColorHistogram, Compactness, Eccentricity, ExtremalPoints, FCTH, FuzzyHistogram, FuzzyOpponentHistogram, Gabor, Haralick, Histogram, JCD, JpegCoefficientHistogram, LocalBinaryPatterns, LuminanceLayout, MeanIntensityLocalBinaryPatterns, MeanPatchIntensityHistogram, Moments, MPEG7ColorLayout, MPEG7EdgeHistogram, OpponentHistogram, PHOG, PolygonEvolution, Profiles, ReferenceColorSimilarity, Sift, SquareModelShapeMatrix, SURF, Tamura, Thumbnail

public abstract class AbstractFeatureDescriptor
extends Object
implements FeatureDescriptor

This abstract class provides some convenient base functionalities for feature descriptors like the getter for the data (including null check) and the propery change support.

Author:

Franz

Nested Class Summary

Nested classes/interfaces inherited from interface de.lmu.ifi.dbs.jfeaturelib.Descriptor

Descriptor.Supports

Field Summary

Fields

Modifier and Type	Field and Description
protected PropertyChangeSupport	pcs Property change support that can be used by the implementing class to inform listeners about updates.

Fields inherited from interface de.lmu.ifi.dbs.jfeaturelib.Descriptor

DOES_ALL

Constructor Summary

Constructors

Constructor and Description
AbstractFeatureDescriptor()

Method Summary

Methods

Modifier and Type	Method and Description
protected void	addData(double[] data) Adds the double array to the list.
protected void	addData(int[] data) Converts the given int array to double[] and adds this array to the list.
protected void	addData(List<double[]> data) Adds the double array list to the list of data arrays.
void	addPropertyChangeListener(PropertyChangeListener listener) Adds a Property change listener for this feature vector.
protected void	endProgress()
protected void	firePropertyChange(Progress event) Propagates the given progress event using property change support.
List<double[]>	getFeatures() Returns a reference to the data calculated by the according descriptor.
ij.process.ImageProcessor	getMask() returns the current mask
protected boolean	inMask(int x, int y) Check wether a pixel is inside a set mask.
protected void	setMask(ij.process.ImageProcessor ip) Sets / stores the mask applied to the referenced image processor.
void	setProperties(LibProperties properties) Injects the properties class that can be used to retrieve several properties.
protected void	startProgress()
EnumSet<Descriptor.Supports>	supports() Determine the capabilities of this algorithm.

Methods inherited from class java.lang.Object

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

Methods inherited from interface de.lmu.ifi.dbs.jfeaturelib.features.FeatureDescriptor

getDescription

Methods inherited from interface de.lmu.ifi.dbs.jfeaturelib.Descriptor

run

Field Detail

pcs

protected final PropertyChangeSupport pcs

Property change support that can be used by the implementing class to inform listeners about updates.

Constructor Detail

AbstractFeatureDescriptor

public AbstractFeatureDescriptor()

Method Detail

getFeatures

public List<double[]> getFeatures()

Returns a reference to the data calculated by the according descriptor. The list will most likely just containa single double array holding the computed values. In cases where a descriptor computes mutliple features (for example SIFT, where a vector is calculated for each point of interest), the list will contain several double arrays.

Specified by:

getFeatures in interface FeatureDescriptor

Returns:

list of feature vectors.

addPropertyChangeListener

public void addPropertyChangeListener(PropertyChangeListener listener)

Adds a Property change listener for this feature vector. During the computation of the descriptor, at least 2 progress events should be fired (Start/End).

Specified by:

addPropertyChangeListener in interface Descriptor

Parameters:

listener -

See Also:

Progress.START, Progress.END, Progress

addData

protected void addData(double[] data)

Adds the double array to the list. Keep in mind that the array is NOT copied but directly put into the list.

See Also:

data

addData

protected void addData(int[] data)

Converts the given int array to double[] and adds this array to the list.

See Also:

data

addData

protected void addData(List<double[]> data)

Adds the double array list to the list of data arrays. Keep in mind that the array is NOT copied but directly put into the list.

Parameters:

data -

See Also:

data

firePropertyChange

protected void firePropertyChange(Progress event)

Propagates the given progress event using property change support. The old value of the firePropertyChange is the most recent progress event that was propagated by this method (null in case of the first event).

Parameters:

event -

setProperties

public void setProperties(LibProperties properties)
                   throws IOException

Description copied from interface: FeatureDescriptor

Injects the properties class that can be used to retrieve several properties.

Specified by:

setProperties in interface FeatureDescriptor

Throws:

IOException - if something went wrong

supports

public EnumSet<Descriptor.Supports> supports()

Description copied from interface: Descriptor

Determine the capabilities of this algorithm.

Specified by:

supports in interface Descriptor

Returns:

the set of supported capabilities.

startProgress

protected void startProgress()

endProgress

protected void endProgress()

setMask

protected void setMask(ij.process.ImageProcessor ip)

Sets / stores the mask applied to the referenced image processor.

Parameters:

ip - the image processor from which the mask should be taken

inMask

protected boolean inMask(int x,
             int y)

Check wether a pixel is inside a set mask. If no mask is applied, then the method always returns true. If a pixel is in the mask, this means that it should be processed.

Parameters:

x -

y -

Returns:

true if the pixel is in the mask and should thus be processed.

getMask

public ij.process.ImageProcessor getMask()

returns the current mask

Returns: