问题描述

单个模板的多目标识别 OPENCV

初学图像处理， 请多多指教！

我现在想解决的是用单个模板去匹配测试图中的多个目标（目标有仿射变换），参照了现成的一些SURF算法提取特征值进行匹配，可是这样只能匹配到一个目标。
后来发现匹配的knnmatch函数可以对一个 query descriptor返回k个最佳匹配，可是效果并不好， 不知有什么建议？十分感谢！

代码如下：

  Mat img1 = imread( ""q.jpg"" CV_LOAD_IMAGE_GRAYSCALE );  Mat img2 = imread( ""qw.jpg"" CV_LOAD_IMAGE_GRAYSCALE );    //-- Step 1: Detect the keypoints using Sift Detector  SiftFeatureDetector  detector( hessian );  vector<KeyPoint> queryKeypoints trainKeypoints;  detector.detect( img1 queryKeypoints );  detector.detect( img2 trainKeypoints );    //-- Step 2: Extract the keypoints using Sift Extractor    Mat queryDescriptortrainDescriptor;// extract keypoints     SiftDescriptorExtractor extractor;  //Create Descriptor Extractor    extractor.compute( img1 queryKeypoints queryDescriptor );    extractor.compute( img2 trainKeypoints trainDescriptor );        //--Step3: Match         vector<vector<DMatch>> m_knnMatches;        vector<DMatch>m_Matches;        vector<DMatch>n_Matches;    const float minRatio = 1.f / 1.5f;        matcher.knnMatch(queryDescriptortrainDescriptorm_knnMatches3);        for (size_t i=0; i<m_knnMatches.size(); i++)        {            const cv::DMatch& bestMatch = m_knnMatches[i][0];            const cv::DMatch& betterMatch = m_knnMatches[i][1];            const cv::DMatch& betterrMatch = m_knnMatches[i][2];            float distanceRatio = bestMatch.distance / betterMatch.distance;            float distanceRatio1 = betterMatch.distance / betterrMatch.distance;            if (distanceRatio < minRatio)        {                m_Matches.push_back(bestMatch);            }            if (distanceRatio1 < minRatio)            {                n_Matches.push_back(betterMatch);            }        }        Mat img_matches;  drawMatches( img1queryKeypointsimg2trainKeypointsm_Matches img_matches Scalar::all(-1) Scalar::all(-1) vector<char>() DrawMatchesFlags::NOT_DRAW_SINGLE_POINTS );        Mat img_matches2;  drawMatches( img1queryKeypointsimg2trainKeypointsn_Matches img_matches2 Scalar::all(-1) Scalar::all(-1) vector<char>() DrawMatchesFlags::NOT_DRAW_SINGLE_POINTS );

解决方案

我记得是可以匹配多个的哇，你代码发上来瞅瞅啊，opencv那些函数效果都一般，你得自己多调试些参数

解决方案二：
surf是尺度不变的啊，你这个有仿射变换问题应该也不是很大，你这个knn就找前三个？是不是有点少了
效果不好可能是特征点找多少了，先用高通滤波器滤波一下，我暂时想到这么多

解决方案三：
特征点这块我也不是很熟，类似有这种是添加一些条件限制什么的，来筛选，距离啊，你说的knn啊什么的

 //-- Quick calculation of max and min distances between keypoints      for( int i = 0; i < descriptors_1.rows; i++ )      {           double dist = matches[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.      std::vector< DMatch > good_matches;      for( int i = 0; i < descriptors_1.rows; i++ )      {           if( matches[i].distance < 0.6*max_dist )          {               good_matches.push_back( matches[i]);           }      }      Mat img_matches;      drawMatches(img_1 keyPoints_1 img_2 keyPoints_2          good_matches img_matches Scalar::all(-1) Scalar::all(-1)          vector<char>() DrawMatchesFlags::NOT_DRAW_SINGLE_POINTS);