OpenCV单目视觉定位（测量）C++代码

#include <opencv2/opencv.hpp> #include <iostream> using namespace std; using namespace cv; //全局变量 Mat src, gray, gray_blur, contours_image,dstThreshold; //本相机分辨率为640*480，定义图像中心为图像原点，也即对应相机的中心位置，光轴位置 //自定义图像原点坐标 float oriX = 296.0f; float oriY = 241.0f; float targetImage_X, targetImage_Y; //目标物距图像原点的X,Y方向的像素距离 float mm_per_pixel; //像素尺寸 float targetLength=10.45f; //目标物实际长度 float targetActualX, targetActualY; //二维空间实际数据 //视觉定位函数 void Location(); int main() { cout <<" 。。。。。。。。。。。。。。。。。按'q'退出程序。。。。。。。。。。。。。。。" << endl; VideoCapture cap(0); if (!cap.isOpened()){ cout << "Failed To Open capture"; return -1; } while (1){ cap >> src; cvtColor(src, gray, CV_BGR2GRAY); imshow("gray", gray); contours_image = Mat::zeros(src.rows, src.cols, CV_8UC3); gray_Contrast = Mat::zeros(gray.size(), gray.type()); Location(); char key = waitKey(10); if (key == 'q'){ break; } } } //定位函数 void Location(){ //均值滤波 blur(gray, gray_blur, Size(3, 3)); //边缘检测提取边缘信息 Canny(gray_blur, dstThreshold, 150, 450); imshow("canny边缘检测", dstThreshold); //对边缘图像提取轮廓信息 vector<vector<Point> >contours; findContours(dstThreshold, contours, CV_RETR_LIST, CV_CHAIN_APPROX_SIMPLE); //画出轮廓 drawContours(contours_image, contours, -1, Scalar(0, 0, 255)); imshow("contours", contours_image); //画出定义的原点 circle(src, Point2f(oriX, oriY), 2, Scalar(0, 0, 255), 3); //定义分别逼近六边形和五边形的轮廓 vector< vector<Point> > Contour1_Ok, Contour2_Ok; //轮廓分析 vector<Point> approx; for (int i = 0; i < contours.size(); i++){ approxPolyDP(Mat(contours[i]), approx, cv::arcLength(cv::Mat(contours[i]), true)*0.04, true); //去除 小轮廓，只提取凸轮廓 if (std::fabs(cv::contourArea(contours[i])) < 600 || !cv::isContourConvex(approx)) continue; //保存逼近六边形的轮廓 到 Contour1_Ok if (approx.size() == 6){ Contour1_Ok.push_back(contours[i]); } //保存逼近五边形的轮廓 到 Contour2_Ok else if (approx.size() == 5){ Contour2_Ok.push_back(contours[i]); } } //对所有符合要求的六边形，五边形轮廓进行分析 //识别出自定义的物体的关键是： //1.六边形和五边形轮廓的最小外接矩形的中心基本在同一点 //2.六边形轮廓的最小外接矩形的任一边长大于五边形轮廓的最小外接矩形的任一边长 for (int i = 0; i < Contour1_Ok.size(); i++){ for (int j = 0; j < Contour2_Ok.size(); j++){ RotatedRect minRect1 = minAreaRect(Mat(Contour1_Ok[i])); //六边形轮廓的最小外接矩形 RotatedRect minRect2 = minAreaRect(Mat(Contour2_Ok[j])); //五边形轮廓的最小外界矩形 //找出符合要求的轮廓的最小外接矩形 if ( fabs(minRect1.center.x - minRect2.center.x) < 30 && fabs(minRect1.center.y - minRect2.center.y)<30 && minRect1.size.width > minRect2.size.width){ Point2f vtx[4]; minRect1.points(vtx); //画出找到的物体的最小外接矩形 for (int j = 0; j < 4; j++) line(src, vtx[j], vtx[(j + 1) % 4], Scalar(0, 0, 255), 2, LINE_AA); //画出目标物中心到图像原点的直线 line(src, minRect1.center, Point2f(oriX, oriY), Scalar(0, 255, 0), 1, LINE_AA); //目标物距图像原点的X,Y方向的像素距离 targetImage_X = minRect1.center.x - oriX; targetImage_Y = oriY - minRect1.center.y; line(src, minRect1.center, Point2f(minRect1.center.x, oriY), Scalar(255, 0, 0), 1, LINE_AA); line(src, Point2f(oriX, oriY), Point2f(minRect1.center.x, oriY), Scalar(255, 0, 0), 1, LINE_AA); Point2f pointX((oriX + minRect1.center.x) / 2, oriY); Point2f pointY(minRect1.center.x, (oriY + minRect1.center.y) / 2); //找出最大边 float a = minRect1.size.height, b = minRect1.size.width; if (a < b) a = b; mm_per_pixel = targetLength / a; //计算像素尺寸 = 目标物的实际长度（cm）/ 目标物在图像上的像素长度（pixels） targetActualX = mm_per_pixel *targetImage_X; //计算实际距离X（cm） targetActualY = mm_per_pixel *targetImage_Y; //计算实际距离Y（cm） //打印信息在图片上 String text1 = "X:"+format("%f", targetImage_X); String text2 = "Y:"+format("%f", targetImage_Y); putText(src, text1, pointX, FONT_HERSHEY_SIMPLEX, 0.4, Scalar(0, 0, 255), 1, 8); putText(src, text2, pointY, FONT_HERSHEY_SIMPLEX, 0.4, Scalar(0, 0, 255), 1, 8); String text3 = "Target_X:"+format("%f", targetActualX); String text4 = "Target_Y:"+format("%f", targetActualY); putText(src, text3, Point(10,30), FONT_HERSHEY_SIMPLEX, 0.7, Scalar(0, 0, 255), 1, 8); putText(src, text4, Point(10,60), FONT_HERSHEY_SIMPLEX, 0.7, Scalar(0, 0, 255), 1, 8); } break; } break; } imshow("SRC", src); }