FeatureExtractor.Clustering.KMeans
Class KMeansClustering

java.lang.Object
  FeatureExtractor.Clustering.GenericClustering
      FeatureExtractor.Clustering.KMeans.KMeansClustering

Direct Known Subclasses:

WindsurfKMeansClustering

public abstract class KMeansClustering
extends GenericClustering

Class for the k-means clustering of images.

Field Summary

Fields inherited from class FeatureExtractor.Clustering.GenericClustering

d

Constructor Summary

KMeansClustering(ClusteringDistance d, int n)
Basic constructor.

Method Summary

ClusteringOutput	cluster(GenericFilterOutput image) Applies the k-means clustering algorithm to a given filtered image.
protected abstract java.util.Vector<Pixel>	selectPixels(GenericFilterOutput image) Abstract method to extract the pixels to be clustered from the output of the filter.
java.lang.String	toString() Returns a string representation of the k-means clustering algorithm.

Methods inherited from class java.lang.Object

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

Constructor Detail

KMeansClustering

public KMeansClustering(ClusteringDistance d,
                        int n)

Basic constructor.

Parameters:

d - Clustering distance to be used to compare pixels.

n - Maximum number of clusters to be generated.

Method Detail

toString

public java.lang.String toString()

Returns a string representation of the k-means clustering algorithm.

Specified by:

toString in class GenericClustering

Returns:

A string representation of the k-means clustering algorithm.

selectPixels

protected abstract java.util.Vector<Pixel> selectPixels(GenericFilterOutput image)

Abstract method to extract the pixels to be clustered from the output of the filter.

Parameters:

image - Filtered image to which the clustering algorithm should be applied.

Returns:

Vector of pixels to which the clustering algorithm will be applied.

cluster

public ClusteringOutput cluster(GenericFilterOutput image)

Applies the k-means clustering algorithm to a given filtered image.

Specified by:

cluster in class GenericClustering

Parameters:

image - Filtered image to which the clustering algorithm should be applied.

Returns:

Generic clustering result.