ORB和FREAK局部特征对比

/************************************************************************ * Copyright(c) 2012 Yang Xian * All rights reserved. * * File: main.cpp * Brief: 对比ORB和FREAK特征描述子的算法效果，基于OpenCV2.4.2 * Version: 1.0 * Author: Yang Xian * Email: [email protected] * Date: 2012/07/10 * History: ************************************************************************/ #include <opencv2/core/core.hpp> #include <opencv2/features2d/features2d.hpp> #include <opencv2/highgui/highgui.hpp> #include <opencv2/nonfree/features2d.hpp> #include <opencv2/legacy/legacy.hpp> #include <iostream> #include <vector> using namespace cv; using namespace std; int main(void) { // ORB Mat img_1_Orb = imread("church01.jpg"); Mat img_2_Orb = imread("church02.jpg"); ORB orb; vector<KeyPoint> keyPoints_1_Orb, keyPoints_2_Orb; Mat descriptors_1_Orb, descriptors_2_Orb; double t = (double)getTickCount(); orb(img_1_Orb, Mat(), keyPoints_1_Orb, descriptors_1_Orb); orb(img_2_Orb, Mat(), keyPoints_2_Orb, descriptors_2_Orb); t = ((double)getTickCount() - t)/getTickFrequency(); cout << "ORB time [s]: " << t << endl; BruteForceMatcher<Hamming> matcher_Orb; vector<DMatch> matches_Orb; matcher_Orb.match(descriptors_1_Orb, descriptors_2_Orb, matches_Orb); double max_dist = 0; double min_dist = 100; //-- Quick calculation of max and min distances between keypoints for (int i=0; i<descriptors_1_Orb.rows; i++) { double dist = matches_Orb[i].distance; if (dist < min_dist) min_dist = dist; if(dist > max_dist) max_dist = dist; } printf("-- Max dist : %f \n", max_dist); printf("-- Min dist : %f \n", min_dist); //-- Draw only "good" matches (i.e. whose distance is less than 0.6*max_dist ) //-- PS.- radiusMatch can also be used here. vector<DMatch> good_matches_Orb; for (int i=0; i<descriptors_1_Orb.rows; i++) { if(matches_Orb[i].distance < 0.6*max_dist) { good_matches_Orb.push_back(matches_Orb[i]); } } Mat img_matches_Orb; drawMatches(img_1_Orb, keyPoints_1_Orb, img_2_Orb, keyPoints_2_Orb, good_matches_Orb, img_matches_Orb, Scalar::all(-1), Scalar::all(-1), vector<char>(), DrawMatchesFlags::NOT_DRAW_SINGLE_POINTS); //////////////////////////////////////////////////////////////////////////////////////////////// //////////////////////////////////////////////////////////////////////////////////////////////// // FREAK Mat imgA_Freak = imread("church01.jpg"); Mat imgB_Freak = imread("church02.jpg"); vector<KeyPoint> keypointsA_Freak, keypointsB_Freak; Mat descriptorsA_Freak, descriptorsB_Freak; vector<DMatch> matches_Freak; // DETECTION // Any openCV detector such as SurfFeatureDetector detector_Freak(2000,4); // DESCRIPTOR // Our proposed FREAK descriptor // (roation invariance, scale invariance, pattern radius corresponding to SMALLEST_KP_SIZE, // number of octaves, optional vector containing the selected pairs) // FREAK extractor(true, true, 22, 4, std::vector<int>()); FREAK freak; // MATCHER // The standard Hamming distance can be used such as // BruteForceMatcher<Hamming> matcher; // or the proposed cascade of hamming distance using SSSE3 BruteForceMatcher<Hamming> matcher_Freak; // detect t = (double)getTickCount(); detector_Freak.detect(imgA_Freak, keypointsA_Freak); detector_Freak.detect(imgB_Freak, keypointsB_Freak); t = ((double)getTickCount() - t)/getTickFrequency(); cout << "FREAK detection time [s]: " << t/1.0 << endl; // extract t = (double)getTickCount(); freak.compute(imgA_Freak, keypointsA_Freak, descriptorsA_Freak); freak.compute(imgB_Freak, keypointsB_Freak, descriptorsB_Freak); t = ((double)getTickCount() - t)/getTickFrequency(); cout << "FREAK extraction time [s]: " << t << endl; // match // t = (double)getTickCount(); matcher_Freak.match(descriptorsA_Freak, descriptorsB_Freak, matches_Freak); // t = ((double)getTickCount() - t)/getTickFrequency(); // std::cout << "matching time [s]: " << t << std::endl; max_dist = 0; min_dist = 100; //-- Quick calculation of max and min distances between keypoints for (int i=0; i<descriptorsA_Freak.rows; i++) { double dist = matches_Freak[i].distance; if (dist < min_dist) min_dist = dist; if(dist > max_dist) max_dist = dist; } printf("-- Max dist : %f \n", max_dist); printf("-- Min dist : %f \n", min_dist); //-- Draw only "good" matches (i.e. whose distance is less than 0.7*max_dist ) //-- PS.- radiusMatch can also be used here. vector<DMatch> good_matches_Freak; for (int i=0; i<descriptorsA_Freak.rows; i++) { if(matches_Freak[i].distance < 0.7*max_dist) { good_matches_Freak.push_back(matches_Freak[i]); } } // Draw matches Mat imgMatch_Freak; drawMatches(imgA_Freak, keypointsA_Freak, imgB_Freak, keypointsB_Freak, good_matches_Freak, imgMatch_Freak, Scalar::all(-1), Scalar::all(-1), vector<char>(), DrawMatchesFlags::NOT_DRAW_SINGLE_POINTS); // display imshow("MatchORB", img_matches_Orb); imshow("matchFREAK", imgMatch_Freak); waitKey(0); return 0; }