Surf and Brisk have terrible accuracy

edit

I have been trying both Surf and Brisk and none of them seem to recognize a single object in a photo.

I have tried with a lot of photos like these:

I also have tried with books and other sort of objects but with no result!

Code of brisk and surf implementations below:

   BRISK
const char * PimA="image1.jpg";   // object
const char * PimB="image2.jpg"; // image

cv::Mat GrayA = getMatByName(PimA);
cv::Mat GrayB = getMatByName(PimB);
std::vector<cv::KeyPoint> keypointsA, keypointsB;
cv::Mat descriptorsA, descriptorsB;
//set brisk parameters

int Threshl=10;
int Octaves=4;
float PatternScales=1.0f;
//declare a variable BRISKD of the type cv::BRISK

cv::BRISK  BRISKD(Threshl,Octaves,PatternScales);//initialize algoritm
BRISKD.create("Feature2D.BRISK");

BRISKD.detect(GrayA, keypointsA);
BRISKD.compute(GrayA, keypointsA,descriptorsA);

BRISKD.detect(GrayB, keypointsB);
BRISKD.compute(GrayB, keypointsB,descriptorsB);


cv::FlannBasedMatcher matcher(new cv::flann::LshIndexParams(20,10,2));


std::vector<cv::DMatch> matches;
matcher.match(descriptorsA, descriptorsB, matches);


std::vector<cv::DMatch> goodMatches;
int min = 1000, max = 0;
for(int i=0; i<matches.size(); i++){
   if(matches[i].distance < min)
       min = matches[i].distance;
   if(matches[i].distance > max)
        max = matches[i].distance;
}

printf("min - max %d %d\n", min, max);

for(int i=0; i<matches.size(); i++)
   if(matches[i].distance < min*3)
     goodMatches.push_back(matches[i]);

cv::Mat all_matches;
cv::drawMatches( GrayA, keypointsA, GrayB, keypointsB,
                 goodMatches, all_matches, cv::Scalar::all(-1), cv::Scalar::all(-1),
                 vector<char>(),cv::DrawMatchesFlags::DEFAULT );
showImage(all_matches, 1);




std::vector<Point2f> obj;
std::vector<Point2f> scene;

for( int i = 0; i < goodMatches.size(); i++ )
{
    //-- Get the keypoints from the good matches
        obj.push_back( keypointsA[ goodMatches[i].queryIdx ].pt );
     scene.push_back( keypointsB[ goodMatches[i].trainIdx ].pt );
}


Mat H = findHomography(obj, scene, CV_RANSAC);

std::vector<Point2f> obj_corners(4);
obj_corners[0] = cvPoint(0,0);
obj_corners[1] = cvPoint( GrayA.cols, 0 );
obj_corners[2] = cvPoint( GrayA.cols, GrayA.rows );
obj_corners[3] = cvPoint( 0, GrayA.rows );
std::vector<Point2f> scene_corners(4);

perspectiveTransform( obj_corners, scene_corners, H);

//-- Draw lines between the corners (the mapped object in the scene - image_2 )
line( GrayB, scene_corners[0] + Point2f( GrayA.cols, 0), scene_corners[1] + Point2f( GrayA.cols, 0), Scalar(0, 0, 255), 4 );
line( GrayB, scene_corners[1] + Point2f( GrayA.cols, 0), scene_corners[2] + Point2f( GrayA.cols, 0), Scalar( 0, 0, 255), 4 );
line( GrayB, scene_corners[2] + Point2f( GrayA.cols, 0), scene_corners[3] + Point2f( GrayA.cols, 0), Scalar( 0, 0, 255), 4 );
line( GrayB, scene_corners[3] + Point2f( GrayA.cols, 0), scene_corners[0] + Point2f( GrayA.cols, 0), Scalar( 0, 0, 255), 4 );


showImage(GrayB, 1);






 SURF    

  Mat img_object = imread( "image1.jpg", CV_LOAD_IMAGE_GRAYSCALE );
  Mat img_scene = imread( "image2.jpg", CV_LOAD_IMAGE_GRAYSCALE );

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

  //-- Step 1: Detect the keypoints using SURF Detector
  int minHessian = 400;

  SurfFeatureDetector detector( minHessian );

  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)
  SurfDescriptorExtractor extractor;

  Mat descriptors_object, descriptors_scene;

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

  //-- Step 3: Matching descriptor vectors using FLANN matcher
  FlannBasedMatcher matcher;
  std::vector< ...