FLANN matcher-OPENCV 3-min max disatance

Question

I have some code which uses the Flann matcher and ORB detector to find features between two images of a person. I'm using opencv 3 on ubuntu. I have a few doubts.. The Code is as follows:

  #include <iostream>
#include </home/sruthi/opencv/include/opencv2/opencv.hpp>

using namespace cv;

//void readme();

/** @function main */
int main(int argc, char** argv)
{
   if( argc != 3 )
  { //readme(); 
     return -1; }

  Mat img_object = imread( argv[1], IMREAD_GRAYSCALE );
  Mat img_scene = imread( argv[2], IMREAD_GRAYSCALE );


    if (!img_object.data || !img_scene.data)
    {
        std::cout << " --(!) Error reading images " << std::endl; return -1;
    }

    //-- Step 1: Detect the keypoints using ORB Detector
    Ptr<FeatureDetector> detector = ORB::create();

    std::vector<KeyPoint> keypoints_object, keypoints_scene;

    detector->detect(img_object, keypoints_object);
    detector->detect(img_scene, keypoints_scene);

    //-- Step 2: Calculate descriptors (feature vectors)
    Ptr<DescriptorExtractor> extractor = ORB::create();

    Mat descriptors_object, descriptors_scene;

    extractor->compute(img_object, keypoints_object, descriptors_object);
    extractor->compute(img_scene, keypoints_scene, descriptors_scene);

    descriptors_object.convertTo(descriptors_object,CV_32F);
    descriptors_scene.convertTo(descriptors_scene,CV_32F);


    //-- Step 3: Matching descriptor vectors using FLANN matcher
    Ptr<DescriptorMatcher> matcher = DescriptorMatcher::create("FlannBased");
    std::vector< DMatch > matches;
    matcher->match(descriptors_object, descriptors_scene, matches);

    double max_dist = 0; double min_dist = 100;

    //-- Quick calculation of max and min distances between keypoints
    for (int i = 0; i < descriptors_object.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 3*min_dist )
    std::vector< DMatch > good_matches;

    for (int i = 0; i < descriptors_object.rows; i++)
    {
        if (matches[i].distance < 3 * min_dist)
        {
            good_matches.push_back(matches[i]);
        }
    }

    Mat img_matches;

    drawMatches(img_object, keypoints_object, img_scene, keypoints_scene, good_matches, img_matches, Scalar::all(-1), Scalar::all(-1), std::vector<char>(), DrawMatchesFlags::NOT_DRAW_SINGLE_POINTS);

    //-- Localize the object
    std::vector<Point2f> obj;
    std::vector<Point2f> scene;
    for (int i = 0; i < good_matches.size(); i++)
    {
        //-- Get the keypoints from the good matches
        obj.push_back(keypoints_object[good_matches[i].queryIdx].pt);
        scene.push_back(keypoints_scene[good_matches[i].trainIdx].pt);
    }


    //-- Show detected matches
  imshow( "Good Matches", img_matches );

  for( int i = 0; i < (int)good_matches.size(); i++ )
  { printf( "-- Good Match [%d] Keypoint 1: %d  -- Keypoint 2: %d  \n", i, good_matches[i].queryIdx, good_matches[i].trainIdx ); }

    waitKey(0);
    return 0;


}

double max_dist = 0; double min_dist = 100; Why do we declare these distances as 0 and 100 respectively? As output I get Max dist : 488.559113 Min dist : 100.000000. It seems wrong.
if (matches[i].distance < 3 * min_dist) { good_matches.push_back(matches[i]); } Why am I not able to change 3*min_dist to 2*min_dist? I get no matches if I do it.
Is it compulsory that there have to parallel lines as shown in the official documentation. I dont get parallel lines. I'm not able to post a screenshot.

Him Him · Accepted Answer · 2015-12-05T23:07:18

These initializations seem to be tuned to the particular case for which this code is written.
This is again due to the tuned parameters. Try experimenting with different set of images and you should see some change in results. Also, another good practice is to use second best distances for finding good matches.
The correct matches should have parallel lines, because the relative position of matches in the 2 images should remain the same. Any cross over lines are the incorrect matches.

FLANN matcher-OPENCV 3-min max disatance

1 Answers