opencv学习之SURF特征检测提取演示

// opencv2413_test.cpp : 定义控制台应用程序的入口点。//#include "stdafx.h"#include <opencv2\opencv.hpp>#include <iostream>#include <string>#include<opencv2\core\core.hpp>#include<opencv2\features2d\features2d.hpp>#include<opencv2\nonfree\nonfree.hpp>using namespace cv;using namespace std;int main(){system("color 2F");Mat srcImage1 = imread("1.jpg",-1);Mat srcImage2 = imread("2.jpg",-1);imshow("src1", srcImage1);imshow("src2", srcImage2);//定义SURF中的hessian阈值特征点检测算子int minHessian = 400;//定义一个SurfFeatureDetector特征检测类对象SurfFeatureDetector detector(minHessian);//定义模板类，vector是可以存放任意类型的动态数组，能够增加和压缩数据vector<KeyPoint> keypoints_1, keypoints_2;//detect函数检测出surf特征关键点，保存在vector容器中detector.detect(srcImage1, keypoints_1);detector.detect(srcImage2, keypoints_2);//绘制特征关键点Mat img_keypoints_1;Mat img_keypoints_2;drawKeypoints(srcImage1, keypoints_1, img_keypoints_1, Scalar::all(-1), DrawMatchesFlags::DRAW_RICH_KEYPOINTS);drawKeypoints(srcImage2, keypoints_2, img_keypoints_2,Scalar::all(-1), DrawMatchesFlags::DRAW_RICH_KEYPOINTS);imshow("特征点检测图1", img_keypoints_1);imshow("特征点检测图2", img_keypoints_2);waitKey();return 0;}

特征提取匹配

// opencv2413_test.cpp : 定义控制台应用程序的入口点。//#include "stdafx.h"#include <opencv2\opencv.hpp>#include <iostream>#include <string>#include<opencv2\core\core.hpp>#include<opencv2\features2d\features2d.hpp>#include<opencv2\nonfree\nonfree.hpp>#include<opencv2\legacy\legacy.hpp>using namespace cv;using namespace std;int main(){system("color 2F");Mat srcImage1 = imread("1.jpg",-1);Mat srcImage2 = imread("2.jpg",-1);imshow("src1", srcImage1);imshow("src2", srcImage2);//定义SURF中的hessian阈值特征点检测算子int minHessian = 400;//定义一个SurfFeatureDetector特征检测类对象SurfFeatureDetector detector(minHessian);//定义模板类，vector是可以存放任意类型的动态数组，能够增加和压缩数据vector<KeyPoint> keypoints_1, keypoints_2;//detect函数检测出surf特征关键点，保存在vector容器中detector.detect(srcImage1, keypoints_1);detector.detect(srcImage2, keypoints_2);//绘制特征关键点Mat img_keypoints_1;Mat img_keypoints_2;drawKeypoints(srcImage1, keypoints_1, img_keypoints_1, Scalar::all(-1), DrawMatchesFlags::DRAW_RICH_KEYPOINTS);drawKeypoints(srcImage2, keypoints_2, img_keypoints_2,Scalar::all(-1), DrawMatchesFlags::DRAW_RICH_KEYPOINTS);imshow("特征点检测图1", img_keypoints_1);imshow("特征点检测图2", img_keypoints_2);/*进行SURF的特征提取*///计算描述符（特征向量）SurfDescriptorExtractor extractor;Mat descriptors1, descriptors2;extractor.compute(srcImage1, keypoints_1, descriptors1);extractor.compute(srcImage2, keypoints_2, descriptors2);//实例化一个匹配器BruteForceMatcher<L2<float>> matcher;vector<DMatch> matches;matcher.match(descriptors1, descriptors2, matches);//绘制从两个图像中匹配出的关键点Mat imgMatches;drawMatches(srcImage1, keypoints_1, srcImage2, keypoints_2, matches, imgMatches);imshow("匹配图", imgMatches);waitKey();return 0;}

使用Flann进行关键点的描述和匹配

// opencv2413_test.cpp : 定义控制台应用程序的入口点。//#include "stdafx.h"#include <opencv2\opencv.hpp>#include <iostream>#include <string>#include<opencv2\core\core.hpp>#include<opencv2\features2d\features2d.hpp>#include<opencv2\nonfree\nonfree.hpp>#include<opencv2\legacy\legacy.hpp>using namespace cv;using namespace std;int main(){system("color 2F");Mat srcImage1 = imread("1.jpg",-1);Mat srcImage2 = imread("2.jpg",-1);imshow("src1", srcImage1);imshow("src2", srcImage2);//定义SURF中的hessian阈值特征点检测算子int minHessian = 400;//定义一个SurfFeatureDetector特征检测类对象SurfFeatureDetector detector(minHessian);//定义模板类，vector是可以存放任意类型的动态数组，能够增加和压缩数据vector<KeyPoint> keypoints_1, keypoints_2;//detect函数检测出surf特征关键点，保存在vector容器中detector.detect(srcImage1, keypoints_1);detector.detect(srcImage2, keypoints_2);//绘制特征关键点Mat img_keypoints_1;Mat img_keypoints_2;drawKeypoints(srcImage1, keypoints_1, img_keypoints_1, Scalar::all(-1), DrawMatchesFlags::DRAW_RICH_KEYPOINTS);drawKeypoints(srcImage2, keypoints_2, img_keypoints_2,Scalar::all(-1), DrawMatchesFlags::DRAW_RICH_KEYPOINTS);imshow("特征点检测图1", img_keypoints_1);imshow("特征点检测图2", img_keypoints_2);/*进行SURF的特征提取*///计算描述符（特征向量）SurfDescriptorExtractor extractor;Mat descriptors1, descriptors2;extractor.compute(srcImage1, keypoints_1, descriptors1);extractor.compute(srcImage2, keypoints_2, descriptors2);/*//实例化一个匹配器BruteForceMatcher<L2<float>> matcher;vector<DMatch> matches;matcher.match(descriptors1, descriptors2, matches);//绘制从两个图像中匹配出的关键点Mat imgMatches;drawMatches(srcImage1, keypoints_1, srcImage2, keypoints_2, matches, imgMatches);imshow("SURF--匹配图", imgMatches);*//*使用Flann进行特征点匹配*/FlannBasedMatcher matcher;vector<DMatch> matches2;matcher.match(descriptors1, descriptors2, matches2);double max_dist = 0;double min_dist = 100;//快速计算关键点之间的最大和最小距离for (int i = 0; i < descriptors1.rows; i++) {double dist = matches2[i].distance;if (dist < min_dist)min_dist = dist;if (dist > max_dist)max_dist = dist;}//输出距离信息cout << "最大距离(Max Dist):" << "\n" << max_dist;cout << "最小距离(Min Dist):" << "\n" << min_dist;//存下所有符合条件的匹配结果，距离小于2*min_distvector<DMatch> good_matches;for (int i = 0; i < descriptors1.rows; i++) {if (matches2[i].distance < 2 * min_dist) {good_matches.push_back(matches2[i]);}}//绘制符合条件的匹配点Mat img_matches;drawMatches(srcImage1, keypoints_1, srcImage2, keypoints_2, good_matches, img_matches, Scalar::all(-1), Scalar::all(-1), vector<char>(), DrawMatchesFlags::NOT_DRAW_SINGLE_POINTS);//输出相关匹配点信息for (int i = 0; i < good_matches.size(); i++) {printf(">符合条件的匹配点[%d] 特征点1：%d -- 特征点2：%d\n", i, good_matches[i].queryIdx, good_matches[i].trainIdx);}imshow("Flann--匹配效果图", img_matches);waitKey();return 0;}

基于摄像头捕获图片进行特征匹配

// opencv2413_test.cpp : 定义控制台应用程序的入口点。//#include "stdafx.h"#include <opencv2\opencv.hpp>#include <iostream>#include <string>#include<opencv2\core\core.hpp>#include<opencv2\features2d\features2d.hpp>#include<opencv2\nonfree\nonfree.hpp>#include<opencv2\legacy\legacy.hpp>using namespace cv;using namespace std;int main(){system("color 2F");//【0】改变console字体颜色//【1】载入图像、显示并转化为灰度图Mat trainImage = imread("3.jpg"), trainImage_gray;imshow("原始图", trainImage);cvtColor(trainImage, trainImage_gray, CV_BGR2GRAY);//【2】检测Surf关键点、提取训练图像描述符vector<KeyPoint> train_keyPoint;Mat trainDescriptor;SurfFeatureDetector featureDetector(80);featureDetector.detect(trainImage_gray, train_keyPoint);SurfDescriptorExtractor featureExtractor;featureExtractor.compute(trainImage_gray, train_keyPoint, trainDescriptor);//【3】创建基于FLANN的描述符匹配对象FlannBasedMatcher matcher;vector<Mat> train_desc_collection(1, trainDescriptor);matcher.add(train_desc_collection);matcher.train();//【4】创建视频对象、定义帧率VideoCapture cap(0);unsigned int frameCount = 0;//帧数//【5】不断循环，直到q键被按下while (char(waitKey(1)) != 'q'){//<1>参数设置int64 time0 = getTickCount();Mat testImage, testImage_gray;cap >> testImage;//采集视频到testImage中if (testImage.empty())continue;//<2>转化图像到灰度cvtColor(testImage, testImage_gray, CV_BGR2GRAY);//<3>检测S关键点、提取测试图像描述符vector<KeyPoint> test_keyPoint;Mat testDescriptor;featureDetector.detect(testImage_gray, test_keyPoint);featureExtractor.compute(testImage_gray, test_keyPoint, testDescriptor);//<4>匹配训练和测试描述符vector<vector<DMatch> > matches;matcher.knnMatch(testDescriptor, matches, 2);// <5>根据劳氏算法（Lowe's algorithm），得到优秀的匹配点vector<DMatch> goodMatches;for (unsigned int i = 0; i < matches.size(); i++){if (matches[i][0].distance < 0.6 * matches[i][1].distance)goodMatches.push_back(matches[i][0]);}//<6>绘制匹配点并显示窗口Mat dstImage;drawMatches(testImage, test_keyPoint, trainImage, train_keyPoint, goodMatches, dstImage);imshow("匹配窗口", dstImage);//<7>输出帧率信息cout << "当前帧率为：" << getTickFrequency() / (getTickCount() - time0) << endl;}waitKey();return 0;}