#include <opencv
2/opencv
.hpp>
#include <opencv
2/features
2d
.hpp>
#include <opencv
2/calib3d
.hpp>
#include <iostream>
using namespace cv;
using namespace std;
int main
() {
// 1
. 加载左右图像
Mat imgLeft = imread
("left
.jpg"
);
Mat imgRight = imread
("right
.jpg"
);
if
(imgLeft
.empty
() || imgRight
.empty
()) {
cerr << "Error: Could not load image files!" << endl;
return -1;
}
//
2. 检测特征点和计算描述子
Ptr<Feature
2D> detector = ORB::create
();
vector<KeyPoint> keypoints1, keypoints
2;
Mat descriptors1, descriptors
2;
detector->detectAndCompute
(imgLeft, noArray
(), keypoints1, descriptors1
);
detector->detectAndCompute
(imgRight, noArray
(), keypoints
2, descriptors
2);
// 3
. 特征匹配
Ptr<DescriptorMatcher> matcher = DescriptorMatcher::create
("BruteForce-
Hamming"
);
vector<DMatch> matches;
matcher->match
(descriptors1, descriptors
2, matches
);
// 4
. 筛选优质匹配点
vector<DMatch> goodMatches;
double minDist =
100, maxDist = 0;
for
(auto m : matches
) {
if
(m
.distance < minDist
) minDist = m
.distance;
if
(m
.distance > maxDist
) maxDist = m
.distance;
}
for
(auto m : matches
) {
if
(m
.distance <= max
(2 * minDist, 30
.0
)) {
goodMatches
.push_back
(m
);
}
}
// 5
. 提取匹配点坐标
vector<Point
2f> points1, points
2;
for
(auto m : goodMatches
) {
points1
.push_back
(keypoints1[m
.queryIdx]
.pt
);
points
2.push_back
(keypoints
2[m
.trainIdx]
.pt
);
}
// 6
. 计算基础矩阵F
Mat F = findFundamentalMat
(points1, points
2, FM_RANSAC, 3, 0
.99
);
// 7
. 计算核线校正变换
Mat H1, H
2;
stereoRectifyUncalibrated
(points1, points
2, F, imgLeft
.size
(), H1, H
2);
// 8
. 应用校正变换
Mat imgLeftRect, imgRightRect;
warpPerspective
(imgLeft, imgLeftRect, H1, imgLeft
.size
());
warpPerspective
(imgRight, imgRightRect, H
2, imgRight
.size
());
// 9
. 显示结果
Mat canvas;
hconcat
(imgLeftRect, imgRightRect, canvas
);
// 绘制水平线验证核线对齐
for
(int i = 0; i < canvas
.rows; i += 30
) {
line
(canvas, Point
(0, i
), Point
(canvas
.cols, i
), Scalar
(0,
255, 0
), 1
);
}
imshow
("Rectified Images
(Uncalibrated
)", canvas
);
imwrite
("rectified_uncalibrated
.jpg", canvas
);
waitKey
(0
);
return 0;
}这个代码对吗
本文介绍了一种计算两个整数间汉明距离的高效方法，通过位运算符“^”进行异或操作，并利用bin()函数将结果转换为二进制形式，最后统计二进制字符串中‘1’的数量得到汉明距离。
扫一扫
286
被折叠的 条评论
为什么被折叠?
到【灌水乐园】发言
