opencv 基于ORB特征点图像拼接

原图


#include <iostream>
#include <opencv2/opencv.hpp>
#include <opencv.hpp>using namespace cv;
using namespace std;void OptimizeSeam(Mat& img1, Mat& trans, Mat& dst);typedef struct
{Point2f left_top;Point2f left_bottom;Point2f right_top;Point2f right_bottom;
}four_corners_t;four_corners_t corners;void CalcCorners(const Mat& H, const Mat& src)
{double v2[] = { 0, 0, 1 };//左上角double v1[3];//变换后的坐标值Mat V2 = Mat(3, 1, CV_64FC1, v2);  //列向量Mat V1 = Mat(3, 1, CV_64FC1, v1);  //列向量V1 = H * V2;//左上角(0,0,1)cout << "V2: " << V2 << endl;cout << "V1: " << V1 << endl;corners.left_top.x = v1[0] / v1[2];corners.left_top.y = v1[1] / v1[2];//左下角(0,src.rows,1)v2[0] = 0;v2[1] = src.rows;v2[2] = 1;V2 = Mat(3, 1, CV_64FC1, v2);  //列向量V1 = Mat(3, 1, CV_64FC1, v1);  //列向量V1 = H * V2;corners.left_bottom.x = v1[0] / v1[2];corners.left_bottom.y = v1[1] / v1[2];//右上角(src.cols,0,1)v2[0] = src.cols;v2[1] = 0;v2[2] = 1;V2 = Mat(3, 1, CV_64FC1, v2);  //列向量V1 = Mat(3, 1, CV_64FC1, v1);  //列向量V1 = H * V2;corners.right_top.x = v1[0] / v1[2];corners.right_top.y = v1[1] / v1[2];//右下角(src.cols,src.rows,1)v2[0] = src.cols;v2[1] = src.rows;v2[2] = 1;V2 = Mat(3, 1, CV_64FC1, v2);  //列向量V1 = Mat(3, 1, CV_64FC1, v1);  //列向量V1 = H * V2;corners.right_bottom.x = v1[0] / v1[2];corners.right_bottom.y = v1[1] / v1[2];}int main(int argc, char* argv[])
{//Mat imageRight = imread("images/imgR.jpg", 1);    //右图//Mat imageLeft = imread("images/imgL.jpg", 1);    //左图//Mat imageRight = imread("images/SR.jpg", 1);    //右图//Mat imageLeft = imread("images/SL.jpg", 1);    //左图Mat imageRight = imread("images/flowerR.jpg", 1);    //右图Mat imageLeft = imread("images/flowerL.jpg", 1);    //左图//灰度图转换  Mat image_r, image_l;cvtColor(imageRight, image_r, COLOR_BGR2GRAY);cvtColor(imageLeft, image_l, COLOR_BGR2GRAY);//直接从可能重复的区域提取特征点匹配 当前是左右图在拼接处大概有1/3是重复的Rect rect_right = Rect(0, 0, imageRight.cols / 3, imageRight.rows);Rect rect_left = Rect(2*imageLeft.cols/3, 0, (imageLeft.cols/3) -1 , imageLeft.rows);Mat image_r_rect = imageRight(Rect(rect_right));Mat image_l_rect = imageLeft(Rect(rect_left));double start = getTickCount();//提取特征点    Ptr<FeatureDetector> ORBDetector = ORB::create(10000);vector<KeyPoint> keyPoints_r, keyPoints_l;ORBDetector->detect(image_r_rect, keyPoints_r);ORBDetector->detect(image_l_rect, keyPoints_l);//特征点描述，为下边的特征点匹配做准备    Ptr<DescriptorExtractor> ORBDescriptor = ORB::create(10000);Mat imageDesc_r, imageDesc_l;ORBDescriptor->compute(image_r_rect, keyPoints_r, imageDesc_r);ORBDescriptor->compute(image_l_rect, keyPoints_l, imageDesc_l);flann::Index flannIndex(imageDesc_r, flann::LshIndexParams(12, 20, 2), cvflann::FLANN_DIST_HAMMING);vector<DMatch> GoodMatchePoints;Mat macthIndex(imageDesc_l.rows, 2, CV_32SC1), matchDistance(imageDesc_l.rows, 2, CV_32FC1);flannIndex.knnSearch(imageDesc_l, macthIndex, matchDistance, 2, flann::SearchParams());// Lowe's algorithm,获取优秀匹配点for (int i = 0; i < matchDistance.rows; i++){if (matchDistance.at<float>(i, 0) < 0.4 * matchDistance.at<float>(i, 1)){DMatch dmatches(i, macthIndex.at<int>(i, 0), matchDistance.at<float>(i, 0));GoodMatchePoints.push_back(dmatches);}}Mat first_match;//drawMatches(imageLeft, keyPoints_l, imageRight, keyPoints_r, GoodMatchePoints, first_match);drawMatches(image_l_rect, keyPoints_l, image_r_rect, keyPoints_r, GoodMatchePoints, first_match);//namedWindow("first_match ", 2);//imshow("first_match ", first_match);//waitKey();vector<Point2f> imagePoints1, imagePoints2;for (int i = 0; i < GoodMatchePoints.size(); i++){imagePoints2.push_back(keyPoints_l[GoodMatchePoints[i].queryIdx].pt);imagePoints1.push_back(keyPoints_r[GoodMatchePoints[i].trainIdx].pt);}if (imagePoints1.size() <= 10 || imagePoints2.size() <= 10){printf("There is little keypoints\n");getchar();}//将左图的坐标转化到原图的位置，否则其变换矩阵在x方向的平稳不对for (auto iter = imagePoints2.begin(); iter != imagePoints2.end(); iter++){(*iter).x  += 2 * imageLeft.cols / 3;}//获取图像1到图像2的投影映射矩阵 尺寸为3*3  Mat homo = findHomography(imagePoints1, imagePoints2, RANSAC);// 也可以使用getPerspectiveTransform方法获得透视变换矩阵，不过要求只能有4个点，效果稍差//Mat   homo=getPerspectiveTransform(imagePoints1,imagePoints2);  cout << "变换矩阵为：\n" << homo << endl << endl; //输出映射矩阵      //计算配准图的四个顶点坐标CalcCorners(homo, imageRight);//cout << "left_top:" << corners.left_top << endl;//cout << "left_bottom:" << corners.left_bottom << endl;//cout << "right_top:" << corners.right_top << endl;//cout << "right_bottom:" << corners.right_bottom << endl;//图像配准  Mat imageTransform1, imageTransform2;warpPerspective(imageRight, imageTransform1, homo, Size(MAX(corners.right_top.x, corners.right_bottom.x), imageLeft.rows));rectangle(imageRight, Rect(imageRight.cols - MAX(corners.right_top.x, corners.right_bottom.x), 0, MAX(corners.right_top.x, corners.right_bottom.x), 500), (0, 0, 255));imshow("rectangle", imageRight);//imshow("透视矩阵变换Right", imageTransform1);//imwrite("trans1.jpg", imageTransform1);//waitKey();//创建拼接后的图,需提前计算图的大小int move_x = 900;int dst_width = imageTransform1.cols;  //取最右点的长度为拼接图的长度int dst_height = imageLeft.rows;Mat dst(dst_height, dst_width, CV_8UC3);dst.setTo(0);imageTransform1.copyTo(dst(Rect(0, 0, imageTransform1.cols, imageTransform1.rows)));imshow("transform1", dst);imageLeft.copyTo(dst(Rect(0, 0, imageLeft.cols, imageLeft.rows)));imshow("b_dst", dst);waitKey();OptimizeSeam(imageLeft, imageTransform1, dst);double end = getTickCount();double useTime = (end - start) / getTickFrequency();cout << "use-time : " << useTime << "s" << endl;imshow("dstOptimize", dst);//imwrite("dst.jpg", dst);waitKey();return 0;
}//优化两图的连接处，使得拼接自然
void OptimizeSeam(Mat& img1, Mat& trans, Mat& dst)
{int start = MIN(corners.left_top.x, corners.left_bottom.x);//开始位置，即重叠区域的左边界  double processWidth = img1.cols - start;//重叠区域的宽度  int rows = dst.rows;int cols = img1.cols; //注意，是列数*通道数double alpha = 1;//img1中像素的权重  for (int i = 0; i < rows; i++){uchar* p = img1.ptr<uchar>(i);  //获取第i行的首地址uchar* t = trans.ptr<uchar>(i);uchar* d = dst.ptr<uchar>(i);for (int j = start; j < cols; j++){//如果遇到图像trans中无像素的黑点，则完全拷贝img1中的数据if (t[j * 3] == 0 && t[j * 3 + 1] == 0 && t[j * 3 + 2] == 0){alpha = 1;}else{//img1中像素的权重，与当前处理点距重叠区域左边界的距离成正比，实验证明，这种方法确实好  alpha = (processWidth - (j - start)) / processWidth;}d[j * 3] = p[j * 3] * alpha + t[j * 3] * (1 - alpha);d[j * 3 + 1] = p[j * 3 + 1] * alpha + t[j * 3 + 1] * (1 - alpha);d[j * 3 + 2] = p[j * 3 + 2] * alpha + t[j * 3 + 2] * (1 - alpha);}}}

特征点匹配

拼接结果

opencv 基于ORB特征点图像拼接相关推荐

基于SIFT特征的图像拼接融合（matlab+vlfeat实现）
基于SIFT特征的图像拼接融合(matlab+vlfeat实现) piccolo,之前做的东西,简单整理下,不是做图像方向的,写的不好轻喷主要原理参看SIFT算法详解和SIFT特征匹配算法介绍--寻 ...
opencv 画orb特征点
ORB关键点画特征点 drawKeypoints 函数绘制匹配结果 drawMatches函数画特征点 drawKeypoints 函数前期准备,读取图像,检测ORB特征点 //读取图像 cv ...
基于OpenCV＆ORB和特征匹配的双视频图像拼接（源码＆部署教程）
1.双视频拼接效果展示 2.视频演示 [项目分享]Python基于OpenCV&ORB和特征匹配的双视频图像拼接(源码&部署教程)_哔哩哔哩_bilibili 3.背景随着汽车电子和 ...
Python基于OpenCV＆ORB和特征匹配的双视频图像拼接（源码＆部署教程）
1.双视频拼接效果展示 2.视频演示 [项目分享]Python基于OpenCV&ORB和特征匹配的双视频图像拼接(源码&部署教程)_哔哩哔哩_bilibili 3.背景随着汽车电子和 ...
ORB特征描述原理、python实现及基于opencv实现
写在前面: 黄宁然,看你看过的算法系列*,不如,就到此吧. 参考文献镇楼: [1]汪洋,扫地机器人定位算法设计与嵌入式系统实现 [2]王雯涛,ORB图像特征提取算法的FPGA设计与实现 [3]房亮,基 ...
java opencv orb_OpenCV提取ORB特征并匹配
一.什么是特征? 图像的特征(Feature),是图像上最具代表性的一些点.所谓最具代表性,就是说这些点包含了图像表述的大部分信息.即使旋转.缩放,甚至调整图像的亮度,这些点仍然稳定地存在,不会丢失. ...
opencv与ORB SLAM提取orb特征点比较
在ORB SLAM中,通过四叉树的方式存储关键点,使得图像上特征点分布均匀,便于追踪.此次实验分别用opencv接口和ORB SLAM实现特征点的提取,效果图如下: 实现程序: #include &l ...
视觉SLAM——ORB特征
一.OpenCV 的ORB 特征--slambook2/ch7/orb_ cv.cpp #include <iostream> #include <opencv2/core/core ...
akaze特征匹配怎么去掉不合适的点_单应性矩阵应用基于特征的图像拼接
点击上方蓝字关注我们微信公众号:OpenCV学堂关注获取更多计算机视觉与深度学习知识前言前面写了一篇关于单应性矩阵的相关文章,结尾说到基于特征的图像拼接跟对象检测中单应性矩阵应用场景.得到很多 ...

opencv 基于ORB特征点图像拼接

opencv 基于ORB特征点图像拼接相关推荐

最新文章

热门文章