org.openimaj.ml.clustering.meanshift

Class ExactMeanShift

java.lang.Object

org.openimaj.ml.clustering.meanshift.ExactMeanShift

public class ExactMeanShift
extends Object

Exact mean shift implementation. The mean shift procedure is applied to every underlying point. This can be quite slow, especially with many points.

Author:

Jonathon Hare (jsh2@ecs.soton.ac.uk)

Constructor Summary

Constructors

Constructor and Description
ExactMeanShift(MultivariateKernelDensityEstimate kde) Perform the ExactMeanShift operation on the given KDE.

Method Summary

Modifier and Type	Method and Description
protected boolean	computeMeanShift(double[] pt)
int[]	getAssignments() Get the assignments
double[][]	getModes() Get the modes
protected void	mergeModes(double[][] modePerPoint)
protected void	performMeanShift()

Methods inherited from class java.lang.Object

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

Constructor Detail

ExactMeanShift

public ExactMeanShift(MultivariateKernelDensityEstimate kde)

Perform the ExactMeanShift operation on the given KDE.

Parameters:

kde -

Method Detail

performMeanShift

protected void performMeanShift()

getModes

public double[][] getModes()

Get the modes

Returns:

the modes

getAssignments

public int[] getAssignments()

Get the assignments

Returns:

the assignments

mergeModes

protected void mergeModes(double[][] modePerPoint)

computeMeanShift

protected boolean computeMeanShift(double[] pt)