FeatureExtractor.Clustering.Windsurf
Class Mahalanobis

java.lang.Object
  FeatureExtractor.Clustering.ClusteringDistance
      FeatureExtractor.Clustering.Windsurf.Mahalanobis

public class Mahalanobis
extends ClusteringDistance

Class implementing the computation of the Mahalanobis distance.

Constructor Summary

Mahalanobis(ClusteringFeatures[][] data)
Basic constructor.

Method Summary

double	distance(ClusteringFeatures p1, ClusteringFeatures p2) Computes the Mahalanobis distance between two points.
boolean	isSingular() Indicates whether the covariance matrix is singular.

Methods inherited from class java.lang.Object

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

Constructor Detail

Mahalanobis

public Mahalanobis(ClusteringFeatures[][] data)

Basic constructor.

Parameters:

data - Array of WindsurfClusteringFeatures used to compute the inverse covariance matrix.

Method Detail

isSingular

public boolean isSingular()

Indicates whether the covariance matrix is singular.

Returns:

true if the covariance matrix is singular, false otherwise.

distance

public double distance(ClusteringFeatures p1,
                       ClusteringFeatures p2)

Computes the Mahalanobis distance between two points.

Specified by:

distance in class ClusteringDistance

Parameters:

p1 - Points to be compared.

p2 - Points to be compared.

Returns:

Mahalanobis distance value between the two points (the maximum distance value if the covariance matrix is singular).