stereodisc
Class ANNMapper

java.lang.Object
  stereodisc.Analyzer
      stereodisc.ANNAccuracy
          stereodisc.ANNMapper

public class ANNMapper
extends ANNAccuracy

Stereo disk photograph analysis. Uses ANN classifier trained with a training set to classify each pixel in each image in the test set. Copyright (c) 1999-2006, Michael Abramoff. All rights reserved.

Constructor Summary

ANNMapper()

Method Summary

static void	main(java.lang.String[] args)
static float[][]	sampleTrainingFeatures(java.lang.String[] names, int[][] indices) Sample the training features names with indices, for all images.
static float[][]	sampleTrainingFeaturesForLeaveOneOut(java.lang.String[] names, int[][] indices, int exclude) Sample the features in names with indices, for all images except exclude.

Methods inherited from class stereodisc.ANNAccuracy

unitvars, zeromeans

Methods inherited from class stereodisc.Analyzer

analyzeProbabilities, createPixelSet, loadFeatures, pixelsToDataset, showAccuracies

Methods inherited from class java.lang.Object

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

Constructor Detail

ANNMapper

public ANNMapper()

Method Detail

main

public static void main(java.lang.String[] args)

sampleTrainingFeaturesForLeaveOneOut

public static float[][] sampleTrainingFeaturesForLeaveOneOut(java.lang.String[] names,
                                                             int[][] indices,
                                                             int exclude)
                                                      throws java.lang.Exception

Sample the features in names with indices, for all images except exclude.

Parameters:

names - String[] the names of the features

indices - int[][] the indices into the images in names.

exclude - int that contains the image to exclude.

Returns:

float[][] the samples.

Throws:

java.lang.Exception

sampleTrainingFeatures

public static float[][] sampleTrainingFeatures(java.lang.String[] names,
                                               int[][] indices)
                                        throws java.lang.Exception

Sample the training features names with indices, for all images.

Parameters:

names - String[] the names of the features

indices - int[][] the indices into the images in names.

Returns:

float[][] the samples.

Throws:

java.lang.Exception