c++ - BRIEF implementation with OpenCV 2.4.10 -

does know of link brief implementation opencv 2.4? regards.

ps: know such questions not welcome on so, primary focus work have done. there similar question quite received.

one of answers questions suggests generic manner sift, extended brief. here modified code.

#include <opencv2/nonfree/nonfree.hpp>  #include <opencv2/highgui/highgui.hpp>  //using namespace std; using namespace cv;  int main(int argc, char *argv[]) {           mat image = imread("load02.jpg", cv_load_image_grayscale);   cv::initmodule_nonfree();   // create smart pointer sift feature detector.   ptr<featuredetector> featuredetector = featuredetector::create("harris"); // "brief written. changed after answer."   vector<keypoint> keypoints;    // detect keypoints   featuredetector->detect(image, keypoints); // note: featuredetector pointer hence '->'.    //similarly, create smart pointer sift extractor.   ptr<descriptorextractor> featureextractor = descriptorextractor::create("brief");    // compute 128 dimension sift descriptor @ each keypoint.   // each row in "descriptors" correspond sift descriptor each keypoint   mat descriptors;   featureextractor->compute(image, keypoints, descriptors);    // if draw detected keypoint check   mat outputimage;   scalar keypointcolor = scalar(255, 0, 0);     // blue keypoints.   drawkeypoints(image, keypoints, outputimage, keypointcolor, drawmatchesflags::default);    namedwindow("output");   imshow("output", outputimage);    char c = ' ';   while ((c = waitkey(0)) != 'q');  // keep window there until user presses 'q' quit.    return 0;  } 

the issue code gives error: first-chance exception @ 0x00007ffb84698b9c in project2.exe: microsoft c++ exception: cv::exception @ memory location 0x00000071f4fbf8e0.

the error results in function execution breaking. tag says execution resume @ namedwindow("output"); line.

could please fix issue, or suggest new code altogether? thanks.

edit: terminal shows error: assertion failed (!outimage.empty()) in cv::drawkeypoints, file ..\..\..\..opencv\modules\features2d\src\draw.cpp, line 115. next statement code resume remains same, drawkepoints called before it.

in opencv, brief descriptorextractor, not featuredetector. according featuredetector::create, factory method not support "brief" algorithm. in other words, featuredetector::create("brief") returns null pointer , program crashes.

the general steps in feature matching are:

1. find interesting (feature) points in image: featuredetector
2. find way describe points: descriptorextractor
3. try match descriptors (feature vectors) in 2 images: descriptormatcher

brief algorithm step 2. can use other methods, harris, orb, ..., in step 1 , pass result step 2 using brief. besides, sift can used in both step 1 , 2 because algorithm provides methods both steps.

---

here's simple example use brief in opencv. first step, find points looks interesting (key points) in image:

vector<keypoint> detectkeypoints(const mat &image)  {     auto featuredetector = featuredetector::create("harris");     vector<keypoint> keypoints;     featuredetector->detect(image, keypoints);     return keypoints; } 

you can try featuredetector algorithm instead of "harris". next step, compute descriptors key points:

mat computedescriptors(const mat &image, vector<keypoint> &keypoints) {     auto featureextractor = descriptorextractor::create("brief");     mat descriptors;     featureextractor->compute(image, keypoints, descriptors);     return descriptors; } 

you can use algorithm different "brief", too. , can see algorithms in descriptorextractor not same algorithms in featuredetector. last step, match 2 descriptors:

vector<dmatch> matchtwoimage(const mat &descriptor1, const mat &descriptor2) {     auto matcher = descriptormatcher::create("bruteforce");     vector<dmatch> matches;     matcher->match(descriptor1, descriptor2, matches);     return matches; } 

similarly, can try different matching algorithm other "bruteforce". main program, can build application functions:

auto img1 = cv::imread("image1.jpg"); auto img2 = cv::imread("image2.jpg");  auto keypoints1 = detectkeypoints(img1); auto keypoints2 = detectkeypoints(img2);  auto descriptor1 = computedescriptors(img1, keypoints1); auto descriptor2 = computedescriptors(img2, keypoints2);  auto matches = matchtwoimage(descriptor1, descriptor2); 

and use matches vector complete application. if want check results, opencv provides functions draw results of step 1 & 3 in image. example, draw matches in final step:

mat result; drawmatches(img1, keypoints1, img2, keypoints2, matches, result); imshow("result", result); waitkey(0);