001/** 002 * Copyright (c) 2011, The University of Southampton and the individual contributors. 003 * All rights reserved. 004 * 005 * Redistribution and use in source and binary forms, with or without modification, 006 * are permitted provided that the following conditions are met: 007 * 008 * * Redistributions of source code must retain the above copyright notice, 009 * this list of conditions and the following disclaimer. 010 * 011 * * Redistributions in binary form must reproduce the above copyright notice, 012 * this list of conditions and the following disclaimer in the documentation 013 * and/or other materials provided with the distribution. 014 * 015 * * Neither the name of the University of Southampton nor the names of its 016 * contributors may be used to endorse or promote products derived from this 017 * software without specific prior written permission. 018 * 019 * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS" AND 020 * ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED 021 * WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE 022 * DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR 023 * ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES 024 * (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; 025 * LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON 026 * ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT 027 * (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS 028 * SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. 029 */ 030package org.openimaj.demos.sandbox.siftlike; 031 032import org.openimaj.feature.OrientedFeatureVector; 033import org.openimaj.image.DisplayUtilities; 034import org.openimaj.image.FImage; 035import org.openimaj.image.MBFImage; 036import org.openimaj.image.colour.RGBColour; 037import org.openimaj.image.feature.local.descriptor.gradient.GradientFeatureProvider; 038import org.openimaj.image.feature.local.descriptor.gradient.GradientFeatureProviderFactory; 039import org.openimaj.image.feature.local.descriptor.gradient.SIFTFeatureProvider; 040import org.openimaj.image.feature.local.detector.dog.extractor.DominantOrientationExtractor; 041import org.openimaj.image.feature.local.extraction.GradientScaleSpaceImageExtractorProperties; 042import org.openimaj.image.processing.convolution.FImageGradients; 043import org.openimaj.math.geometry.point.Point2dImpl; 044import org.openimaj.math.geometry.shape.Ellipse; 045 046import Jama.Matrix; 047 048public class EllipseGradientFeatureExtractor { 049 GradientFeatureProviderFactory factory; 050 int patchSize = 100; 051 052 public EllipseGradientFeatureExtractor() { 053 this.factory = new SIFTFeatureProvider(); 054 } 055 056 public EllipseGradientFeatureExtractor(GradientFeatureProviderFactory factory) { 057 this.factory = factory; 058 } 059 060 public OrientedFeatureVector[] extract(FImage image, Ellipse ellipse) { 061 final Matrix tf = ellipse.transformMatrix(); 062 final FImage patch = new FImage(patchSize, patchSize); 063 final float halfSize = patchSize / 2; 064 065 // Sample the ellipse content into a rectified image 066 for (int y = 0; y < patchSize; y++) { 067 for (int x = 0; x < patchSize; x++) { 068 final Point2dImpl pt = new Point2dImpl((x - halfSize) / halfSize, (y - halfSize) / halfSize); 069 final Point2dImpl tpt = pt.transform(tf); 070 patch.pixels[y][x] = image.getPixelInterpNative(tpt.x, tpt.y, 0); 071 } 072 } 073 074 // now find grad mags and oris 075 final FImageGradients gmo = FImageGradients.getGradientMagnitudesAndOrientations(patch); 076 077 final GradientScaleSpaceImageExtractorProperties<FImage> props = new GradientScaleSpaceImageExtractorProperties<FImage>(); 078 props.image = patch; 079 props.magnitude = gmo.magnitudes; 080 props.orientation = gmo.orientations; 081 props.x = patch.width / 2; 082 props.y = patch.height / 2; 083 props.scale = patch.height / 2 / 3; // ??? 084 085 final DominantOrientationExtractor doe = new DominantOrientationExtractor(); 086 final float[] oris = doe.extractFeatureRaw(props); 087 088 final MBFImage p2 = patch.toRGB(); 089 for (final float o : oris) { 090 p2.drawLine(p2.getWidth() / 2, p2.getHeight() / 2, o, 20, RGBColour.RED); 091 } 092 DisplayUtilities.display(p2); 093 094 final OrientedFeatureVector[] vectors = new OrientedFeatureVector[oris.length]; 095 for (int i = 0; i < oris.length; i++) { 096 final float ori = oris[i]; 097 final GradientFeatureProvider provider = factory.newProvider(); 098 099 // and construct the feature and sampling every pixel in the patch 100 // note: the descriptor is actually computed over a sub-patch; there 101 // is 102 // a border that is used for oversampling and avoiding edge effects. 103 final float overSample = provider.getOversamplingAmount(); 104 for (int y = 0; y < patchSize; y++) { 105 final float yy = (y * (2 * overSample + 1) / patchSize) - overSample; 106 107 for (int x = 0; x < patchSize; x++) { 108 final float xx = (x * (2 * overSample + 1) / patchSize) - overSample; 109 110 final float gradmag = gmo.magnitudes.pixels[y][x]; 111 final float gradori = gmo.orientations.pixels[y][x]; 112 provider.addSample(xx, yy, gradmag, gradori - ori); 113 } 114 } 115 116 vectors[i] = provider.getFeatureVector(); 117 } 118 119 return vectors; 120 } 121}