/*
* Copyright 2010, 2011 Institut Pasteur.
*
* This file is part of NHerve Main Toolbox, which is an ICY plugin.
*
* NHerve Main Toolbox is free software: you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
* the Free Software Foundation, either version 3 of the License, or
* (at your option) any later version.
*
* NHerve Main Toolbox is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
*
* You should have received a copy of the GNU General Public License
* along with NHerve Main Toolbox. If not, see <http://www.gnu.org/licenses/>.
*/
package plugins.nherve.toolbox.image.feature.com;
import icy.image.IcyBufferedImage;
import icy.type.TypeUtil;
import java.util.List;
import plugins.nherve.toolbox.Algorithm;
import plugins.nherve.toolbox.image.BinaryIcyBufferedImage;
import plugins.nherve.toolbox.image.feature.FeatureException;
import plugins.nherve.toolbox.image.feature.SegmentableIcyBufferedImage;
import plugins.nherve.toolbox.image.feature.descriptor.ColorPixel;
import plugins.nherve.toolbox.image.feature.region.IcyPixel;
import plugins.nherve.toolbox.image.feature.signature.DefaultVectorSignature;
import plugins.nherve.toolbox.image.mask.Mask;
import plugins.nherve.toolbox.image.mask.MaskStack;
import plugins.nherve.toolbox.image.toolboxes.ColorSpaceTools;
/**
* The Class CooccurenceMatrixFactory.
*
* @author Nicolas HERVE - nicolas.herve@pasteur.fr
*/
public class CooccurenceMatrixFactory extends Algorithm {
/** The kernel. */
private List<IcyPixel> kernel;
/**
* Instantiates a new cooccurence matrix factory.
*/
public CooccurenceMatrixFactory() {
super();
setKernel(KernelFactory.getStandardKernel(1));
}
/**
* Gets the index manage borders.
*
* @param data
* the data
* @param px
* the px
* @param w
* the w
* @param h
* the h
* @return the index manage borders
*/
private int getIndexManageBorders(int[] data, IcyPixel px, int w, int h) {
int x = (int)px.x;
int y = (int)px.y;
if (x < 0) {
x = Math.abs(x);
} else if (x >= w) {
x -= 2 * (x - w + 1);
}
if (y < 0) {
y = Math.abs(y);
} else if (y >= h) {
y -= 2 * (y - h + 1);
}
return data[x + y * w];
}
/**
* Gets the indexed image.
*
* @param seg
* the seg
* @return the indexed image
* @throws FeatureException
* the feature exception
*/
public static IcyBufferedImage getIndexedImage(MaskStack seg) throws FeatureException {
final int w = seg.getWidth();
final int h = seg.getHeight();
final int s = seg.size();
IcyBufferedImage index = new IcyBufferedImage(w, h, 1, TypeUtil.TYPE_INT);
int[] idxData = index.getDataXYAsInt(0);
byte[][] segData = new byte[s][];
int id = 0;
for (Mask m : seg) {
segData[id] = m.getBinaryData().getDataXYAsByte(0);
id++;
}
int idx = 0;
for (int y = 0; y < h; y++) {
for (int x = 0; x < w; x++) {
for (id = 0; id < s; id++) {
if (segData[id][idx] == BinaryIcyBufferedImage.TRUE) {
idxData[idx] = id;
break;
}
}
idx++;
}
}
index.dataChanged();
return index;
}
/**
* Builds the from segmentation.
*
* @param seg
* the seg
* @return the cooccurence matrix
* @throws FeatureException
* the feature exception
*/
public CooccurenceMatrix<Integer> buildFromSegmentation(MaskStack seg) throws FeatureException {
return buildFromIndexedImage(getIndexedImage(seg));
}
/**
* Builds the from indexed image.
*
* @param img
* the img
* @param vocabulary
* the vocabulary
* @return the cooccurence matrix
* @throws FeatureException
* the feature exception
*/
public CooccurenceMatrix<Integer> buildFromIndexedImage(IcyBufferedImage img, Vocabulary<Integer> vocabulary) throws FeatureException {
if (img.getDataType() != TypeUtil.TYPE_INT) {
throw new FeatureException("Only TYPE_INT IcyBufferedImage supported in CooccurenceMatrix.buildFromIndexedImage()");
}
if (kernel == null) {
throw new FeatureException("No kernel defined in CooccurenceMatrix.buildFromIndexedImage()");
}
int w = img.getWidth();
int h = img.getHeight();
CooccurenceMatrix<Integer> result = new CooccurenceMatrix<Integer>(vocabulary);
int[] data = img.getDataXYAsInt(0);
for (int x = 0; x < w; x++) {
for (int y = 0; y < h; y++) {
IcyPixel ct = new IcyPixel(x, y);
int center = vocabulary.getIndex(getIndexManageBorders(data, ct, w, h));
for (IcyPixel shift : kernel) {
IcyPixel nb = ct.plus(shift);
int neihbour = vocabulary.getIndex(getIndexManageBorders(data, nb, w, h));
result.add(center, neihbour, 1);
}
}
}
return result;
}
/**
* Builds the from indexed image.
*
* @param idxImg
* the idx img
* @param oriImg
* the ori img
* @param vocabulary
* the vocabulary
* @return the cooccurence matrix
* @throws FeatureException
* the feature exception
*/
public CooccurenceMatrix<Integer> buildFromIndexedImage(IcyBufferedImage idxImg, SegmentableIcyBufferedImage oriImg, VocabularyOfObjects<Integer, DefaultVectorSignature> vocabulary) throws FeatureException {
if (idxImg.getDataType() != TypeUtil.TYPE_INT) {
throw new FeatureException("Only TYPE_INT IcyBufferedImage supported in CooccurenceMatrix.buildFromIndexedImage()");
}
if (kernel == null) {
throw new FeatureException("No kernel defined in CooccurenceMatrix.buildFromIndexedImage()");
}
int w = idxImg.getWidth();
int h = idxImg.getHeight();
ColorPixel colpix = new ColorPixel(ColorSpaceTools.RGB, false);
CooccurenceMatrix<Integer> result = new CooccurenceMatrix<Integer>(vocabulary);
int[] idxData = idxImg.getDataXYAsInt(0);
for (int x = 0; x < w; x++) {
for (int y = 0; y < h; y++) {
IcyPixel ct = new IcyPixel(x, y);
int center = vocabulary.getIndex(getIndexManageBorders(idxData, ct, w, h));
DefaultVectorSignature vs = (DefaultVectorSignature) colpix.extractLocalSignature(oriImg, ct);
vs.multiply(256);
double sct = vocabulary.similarity(center, vs);
for (IcyPixel shift : kernel) {
IcyPixel nb = ct.plus(shift);
DefaultVectorSignature vs2 = (DefaultVectorSignature) colpix.extractLocalSignature(oriImg, nb);
vs2.multiply(256);
int neihbour = vocabulary.getIndex(getIndexManageBorders(idxData, nb, w, h));
double snb = vocabulary.similarity(neihbour, vs2);
result.add(center, neihbour, sct * snb);
}
}
}
return result;
}
/**
* Gets the indexed vocabulary.
*
* @param img
* the img
* @return the indexed vocabulary
* @throws FeatureException
* the feature exception
*/
public static StandardIntegerVocabulary getIndexedVocabulary(IcyBufferedImage img) throws FeatureException {
if (img.getDataType() != TypeUtil.TYPE_INT) {
throw new FeatureException("Only TYPE_INT IcyBufferedImage supported in CooccurenceMatrix.buildFromIndexedImage()");
}
StandardIntegerVocabulary vocabulary = new StandardIntegerVocabulary();
int[] data = img.getDataXYAsInt(0);
for (int d : data) {
if (!vocabulary.contains(d)) {
vocabulary.add(d);
}
}
return vocabulary;
}
/**
* Builds the from indexed image.
*
* @param img
* the img
* @return the cooccurence matrix
* @throws FeatureException
* the feature exception
*/
public CooccurenceMatrix<Integer> buildFromIndexedImage(IcyBufferedImage img) throws FeatureException {
StandardIntegerVocabulary vocabulary = getIndexedVocabulary(img);
return buildFromIndexedImage(img, vocabulary);
}
/**
* Gets the kernel.
*
* @return the kernel
*/
public List<IcyPixel> getKernel() {
return kernel;
}
/**
* Sets the kernel.
*
* @param kernel
* the new kernel
*/
public void setKernel(List<IcyPixel> kernel) {
this.kernel = kernel;
}
/**
* Sets the kernel.
*
* @param kernel
* the new kernel
*/
public void setKernel(int kernel) {
setKernel(KernelFactory.getStandardKernel(kernel));
}
}