OpenCV  3.0.0-rc1
Open Source Computer Vision
 All Classes Namespaces Files Functions Variables Typedefs Enumerations Enumerator Properties Friends Macros Groups Pages
AKAZE local features matching

Introduction

In this tutorial we will learn how to use AKAZE [4] local features to detect and match keypoints on two images. We will find keypoints on a pair of images with given homography matrix, match them and count the

number of inliers (i. e. matches that fit in the given homography).

You can find expanded version of this example here: https://github.com/pablofdezalc/test_kaze_akaze_opencv

Data

We are going to use images 1 and 3 from Graffity sequence of Oxford dataset.

graf.png

Homography is given by a 3 by 3 matrix:

7.6285898e-01 -2.9922929e-01 2.2567123e+02
3.3443473e-01 1.0143901e+00 -7.6999973e+01
3.4663091e-04 -1.4364524e-05 1.0000000e+00

You can find the images (graf1.png, graf3.png) and homography (H1to3p.xml) in opencv/samples/cpp.

Source Code

1 #include <opencv2/features2d.hpp>
2 #include <opencv2/imgcodecs.hpp>
3 #include <opencv2/opencv.hpp>
4 #include <vector>
5 #include <iostream>
6 
7 using namespace std;
8 using namespace cv;
9 
10 const float inlier_threshold = 2.5f; // Distance threshold to identify inliers
11 const float nn_match_ratio = 0.8f; // Nearest neighbor matching ratio
12 
13 int main(void)
14 {
15  Mat img1 = imread("../data/graf1.png", IMREAD_GRAYSCALE);
16  Mat img2 = imread("../data/graf3.png", IMREAD_GRAYSCALE);
17 
18  Mat homography;
19  FileStorage fs("../data/H1to3p.xml", FileStorage::READ);
20  fs.getFirstTopLevelNode() >> homography;
21 
22  vector<KeyPoint> kpts1, kpts2;
23  Mat desc1, desc2;
24 
25  Ptr<AKAZE> akaze = AKAZE::create();
26  akaze->detectAndCompute(img1, noArray(), kpts1, desc1);
27  akaze->detectAndCompute(img2, noArray(), kpts2, desc2);
28 
29  BFMatcher matcher(NORM_HAMMING);
30  vector< vector<DMatch> > nn_matches;
31  matcher.knnMatch(desc1, desc2, nn_matches, 2);
32 
33  vector<KeyPoint> matched1, matched2, inliers1, inliers2;
34  vector<DMatch> good_matches;
35  for(size_t i = 0; i < nn_matches.size(); i++) {
36  DMatch first = nn_matches[i][0];
37  float dist1 = nn_matches[i][0].distance;
38  float dist2 = nn_matches[i][1].distance;
39 
40  if(dist1 < nn_match_ratio * dist2) {
41  matched1.push_back(kpts1[first.queryIdx]);
42  matched2.push_back(kpts2[first.trainIdx]);
43  }
44  }
45 
46  for(unsigned i = 0; i < matched1.size(); i++) {
47  Mat col = Mat::ones(3, 1, CV_64F);
48  col.at<double>(0) = matched1[i].pt.x;
49  col.at<double>(1) = matched1[i].pt.y;
50 
51  col = homography * col;
52  col /= col.at<double>(2);
53  double dist = sqrt( pow(col.at<double>(0) - matched2[i].pt.x, 2) +
54  pow(col.at<double>(1) - matched2[i].pt.y, 2));
55 
56  if(dist < inlier_threshold) {
57  int new_i = static_cast<int>(inliers1.size());
58  inliers1.push_back(matched1[i]);
59  inliers2.push_back(matched2[i]);
60  good_matches.push_back(DMatch(new_i, new_i, 0));
61  }
62  }
63 
64  Mat res;
65  drawMatches(img1, inliers1, img2, inliers2, good_matches, res);
66  imwrite("res.png", res);
67 
68  double inlier_ratio = inliers1.size() * 1.0 / matched1.size();
69  cout << "A-KAZE Matching Results" << endl;
70  cout << "*******************************" << endl;
71  cout << "# Keypoints 1: \t" << kpts1.size() << endl;
72  cout << "# Keypoints 2: \t" << kpts2.size() << endl;
73  cout << "# Matches: \t" << matched1.size() << endl;
74  cout << "# Inliers: \t" << inliers1.size() << endl;
75  cout << "# Inliers Ratio: \t" << inlier_ratio << endl;
76  cout << endl;
77 
78  return 0;
79 }
bool imwrite(const String &filename, InputArray img, const std::vector< int > &params=std::vector< int >())
Saves an image to a specified file.
Definition: base.hpp:172
Mat imread(const String &filename, int flags=IMREAD_COLOR)
Loads an image from a file.
void drawMatches(InputArray img1, const std::vector< KeyPoint > &keypoints1, InputArray img2, const std::vector< KeyPoint > &keypoints2, const std::vector< DMatch > &matches1to2, InputOutputArray outImg, const Scalar &matchColor=Scalar::all(-1), const Scalar &singlePointColor=Scalar::all(-1), const std::vector< char > &matchesMask=std::vector< char >(), int flags=DrawMatchesFlags::DEFAULT)
Draws the found matches of keypoints from two images.
#define CV_64F
Definition: cvdef.h:107
Definition: imgcodecs.hpp:64
int queryIdx
Definition: types.hpp:735
void sqrt(InputArray src, OutputArray dst)
Calculates a square root of array elements.
Brute-force descriptor matcher.
Definition: features2d.hpp:913
InputOutputArray noArray()
XML/YAML file storage class that encapsulates all the information necessary for writing or reading da...
Definition: persistence.hpp:298
Template class for smart pointers with shared ownership.
Definition: cvstd.hpp:283
virtual void detectAndCompute(InputArray image, InputArray mask, std::vector< KeyPoint > &keypoints, OutputArray descriptors, bool useProvidedKeypoints=false)
for i
Definition: modelConvert.m:63
void pow(InputArray src, double power, OutputArray dst)
Raises every array element to a power.
int trainIdx
Definition: types.hpp:736
int main(int argc, const char *argv[])
Definition: facerec_demo.cpp:67
n-dimensional dense array class
Definition: mat.hpp:726
_Tp & at(int i0=0)
Returns a reference to the specified array element.
Class for matching keypoint descriptors.
Definition: types.hpp:728

Explanation

  1. Load images and homography
    Mat img1 = imread("graf1.png", IMREAD_GRAYSCALE);
    Mat img2 = imread("graf3.png", IMREAD_GRAYSCALE);
    Mat homography;
    FileStorage fs("H1to3p.xml", FileStorage::READ);
    fs.getFirstTopLevelNode() >> homography;
    We are loading grayscale images here. Homography is stored in the xml created with FileStorage.
  2. Detect keypoints and compute descriptors using AKAZE
    vector<KeyPoint> kpts1, kpts2;
    Mat desc1, desc2;
    AKAZE akaze;
    akaze(img1, noArray(), kpts1, desc1);
    akaze(img2, noArray(), kpts2, desc2);
    We create AKAZE object and use it's operator() functionality. Since we don't need the mask parameter, noArray() is used.
  3. Use brute-force matcher to find 2-nn matches
    BFMatcher matcher(NORM_HAMMING);
    vector< vector<DMatch> > nn_matches;
    matcher.knnMatch(desc1, desc2, nn_matches, 2);
    We use Hamming distance, because AKAZE uses binary descriptor by default.
  4. Use 2-nn matches to find correct keypoint matches
    for(size_t i = 0; i < nn_matches.size(); i++) {
    DMatch first = nn_matches[i][0];
    float dist1 = nn_matches[i][0].distance;
    float dist2 = nn_matches[i][1].distance;
    if(dist1 < nn_match_ratio * dist2) {
    matched1.push_back(kpts1[first.queryIdx]);
    matched2.push_back(kpts2[first.trainIdx]);
    }
    }
    If the closest match is ratio closer than the second closest one, then the match is correct.
  5. Check if our matches fit in the homography model

    for(int i = 0; i < matched1.size(); i++) {
    Mat col = Mat::ones(3, 1, CV_64F);
    col.at<double>(0) = matched1[i].pt.x;
    col.at<double>(1) = matched1[i].pt.y;
    col = homography * col;
    col /= col.at<double>(2);
    float dist = sqrt( pow(col.at<double>(0) - matched2[i].pt.x, 2) +
    pow(col.at<double>(1) - matched2[i].pt.y, 2));
    if(dist < inlier_threshold) {
    int new_i = inliers1.size();
    inliers1.push_back(matched1[i]);
    inliers2.push_back(matched2[i]);
    good_matches.push_back(DMatch(new_i, new_i, 0));
    }
    }

    If the distance from first keypoint's projection to the second keypoint is less than threshold, then it it fits in the homography.

    We create a new set of matches for the inliers, because it is required by the drawing function.

  6. Output results
    Mat res;
    drawMatches(img1, inliers1, img2, inliers2, good_matches, res);
    imwrite("res.png", res);
    ...
    Here we save the resulting image and print some statistics.

Results

Found matches

res.png

A-KAZE Matching Results

Keypoints 1: 2943
Keypoints 2: 3511
Matches: 447
Inliers: 308
Inlier Ratio: 0.689038}