opencv3.2 SURF实现特征点匹配

opencv3.2中SurfFeatureDetector、SurfDescriptorExtractor、BruteForceMatcher这三个的使用方法已经和原先2.4版本前不一样了。

使用方法示例如下：

 Ptr<SURF> detector = SURF::create(minHessian);
 detector->detect(img_1, keypoints_1);

 Ptr<SURF> extractor = SURF::create();
 extractor->compute(img_1, keypoints_1, descriptors_1);

Ptr<DescriptorMatcher> matcher = DescriptorMatcher::create("BruteForce");//这里填写使用的匹配方式
matcher->match(descriptors_1, descriptors_2, matches);

debug版本detect函数运行时会报错，内存访问错误什么的，，，，，
具体原因还不知道，网上查找资料修改图片type为CV_8U，和给vector手动分配空间，实测没有用，但是改为release版本可以使用。

代码：

#include<opencv2/features2d/features2d.hpp>
#include<opencv2/opencv.hpp>
#include<opencv2/highgui/highgui.hpp>
#include<opencv2/xfeatures2d/nonfree.hpp>
#include<opencv2/core/core.hpp>

#include<iostream>
using namespace std;
using namespace cv;
using namespace cv::xfeatures2d;
int main()
{
    Mat srcImage1 = imread("3.jpg", 1);
    Mat srcImage2 = imread("4.jpg",1);
    if (!srcImage1.data || !srcImage2.data)
    {
        cout << "读取图片出错" << endl;
        return false;
    }

    imshow("原始图1",srcImage1);
    imshow("原始图2", srcImage2);

    int minHessian = 100;
    Ptr<SurfFeatureDetector> detector = SurfFeatureDetector::create(minHessian);

    vector<cv::KeyPoint> key_points_1, key_points_2;

    Mat dstImage1, dstImage2;
    detector->detectAndCompute(srcImage1,Mat(), key_points_1,dstImage1);
    detector->detectAndCompute(srcImage2,Mat(), key_points_2,dstImage2);//可以分成detect和compute

    Mat img_keypoints_1, img_keypoints_2;
    drawKeypoints(srcImage1,key_points_1,img_keypoints_1,Scalar::all(-1),DrawMatchesFlags::DEFAULT);
    drawKeypoints(srcImage2, key_points_2, img_keypoints_2, Scalar::all(-1),DrawMatchesFlags::DEFAULT);

    Ptr<DescriptorMatcher> matcher = DescriptorMatcher::create("FlannBased");
    vector<DMatch>mach;

    matcher->match(dstImage1,dstImage2,mach);
    double Max_dist = 0;
    double Min_dist = 100;
    for (int i = 0; i < dstImage1.rows; i++)
    {
        double dist = mach[i].distance;
        if (dist < Min_dist)Min_dist = dist;
        if (dist > Max_dist)Max_dist = dist;
    }
    cout << "最短距离" << Min_dist << endl;
    cout << "最长距离" << Max_dist << endl;

    vector<DMatch>goodmaches;
    for (int i = 0; i < dstImage1.rows; i++)
    {
        if (mach[i].distance < 2 * Min_dist)
            goodmaches.push_back(mach[i]);
    }
    Mat img_maches;
    drawMatches(srcImage1,key_points_1,srcImage2,key_points_2,goodmaches,img_maches);

    for (int i = 0; i < goodmaches.size(); i++)
    {
        cout << "符合条件的匹配：" << goodmaches[i].queryIdx << "--" << goodmaches[i].trainIdx << endl;
    }
    imshow("效果图1", img_keypoints_1);
    imshow("效果图2", img_keypoints_2);
    imshow("匹配效果",img_maches);

    waitKey(0);
    return 0;
}