车道线检测，尝试用传统的方法识别车道线，可以节约硬件资源，但由于场景太复杂，后面还是采用网络模型识别，不过可以作为预判：
#include <iostream> #include <cmath> #include <vector> #include <algorithm> #include <opencv2/opencv.hpp> using namespace cv; using namespace std; int lane_roi_hulf(cv::Mat& img_gray, cv::Mat& img_hulf); int lane_roi(cv::Mat& img_canny, cv::Mat& img_roi); std::vector<std::vector<cv::Vec4i>> line_separa(std::vector<cv::Vec4i> lines, cv::Mat& img_edges); std::vector<cv::Point> regression(std::vector<std::vector<cv::Vec4i> > left_right_lines, cv::Mat& inputImage); int find_contour(cv::Mat& img_roi, cv::Mat& img_contour); int lane_compel(cv::Mat& img_lane); int main(int argc, char** argv) { cv::Mat img_input, img_guas, img_gray, img_canny; cv::Mat img_hulf, img_roi, img_roi_cont; img_input = cv::imread("./img/city_03_16_229.jpg"); cv::Mat img_uch(img_input.rows, img_input.cols, CV_8UC1, cv::Scalar::all(0)); double time_start = static_cast<double>(getTickCount()); cv::cvtColor(img_input, img_gray, COLOR_BGR2GRAY); //lane_roi_hulf(img_gray, img_hulf); cv::GaussianBlur(img_gray, img_guas, cv::Size(5, 5), 0, 0); cv::threshold(img_guas, img_uch, 0, 255, CV_THRESH_OTSU + CV_THRESH_BINARY); lane_roi(img_uch, img_roi_cont); cv::imshow("img_roi_cont", img_roi_cont); cv::Canny(img_guas, img_canny, 100, 250); cv::imshow("img_canny", img_canny); lane_roi(img_canny, img_roi); //double time = (static_cast<double>(getTickCount()) - time_start) / getTickFrequency(); //std::cout<<"time: "<<time<<std::endl; cv::imshow("img_roi", img_roi); std::vector<cv::Vec4i> lines; HoughLinesP(img_roi, lines, 1, CV_PI / 180, 40, 40, 20); cv::Mat img_contour = cv::Mat::zeros(img_roi.size(), img_roi.type()); find_contour(img_roi_cont, img_contour); imshow("img_contour", img_contour); cv::Mat img_hough = cv::Mat::zeros(img_uch.size(), img_uch.type()); int j = 0 ; cv::Point hough_ini, hough_fini; for(auto i : lines) { hough_ini = cv::Point(i[0], i[1]); hough_fini = cv::Point(i[2], i[3]); cv::line(img_hough, hough_ini, hough_fini, cv::Scalar(255), 1, CV_AA); j++; } std::cout<<"j: "<<j<<std::endl; cv::imshow("img_hough", img_hough); std::vector<std::vector<cv::Vec4i> > left_right_lines; std::vector<cv::Point> lanes; left_right_lines = line_separa(lines, img_roi); lanes = regression(left_right_lines, img_roi); cv::Mat img_lane = cv::Mat::zeros(img_hough.size(), img_hough.type()); int k = 0; cv::Point left_ini, left_fini; for(auto i : left_right_lines[0]) { left_ini = cv::Point(i[0], i[1]); left_fini = cv::Point(i[2], i[3]); cv::line(img_lane, left_ini, left_fini, cv::Scalar(255), 1, CV_AA); k++; } cv::Point right_ini, right_fini; for(auto i : left_right_lines[1]) { right_ini = cv::Point(i[0], i[1]); right_fini = cv::Point(i[2], i[3]); cv::line(img_lane, right_ini, right_fini, cv::Scalar(255), 1, CV_AA); k++; } std::cout<<"k: "<<k<<std::endl; cv::line(img_lane, cv::Point(640, 0), cv::Point(640, 720), cv::Scalar(255), 1, CV_AA); //cv::line(img_lane, lanes[0], lanes[1], cv::Scalar(255), 1, CV_AA); //cv::line(img_lane, lanes[2], lanes[3], cv::Scalar(255), 1, CV_AA); cv::imshow("img_lane", img_lane); cv::waitKey(0); return 0; } int lane_roi(cv::Mat& img_canny, cv::Mat& img_roi) { int rows_kerl = img_canny.rows / 10; int cols_kerl = img_canny.cols / 10; cv::Mat img_fill = cv::Mat::zeros(img_canny.size(), img_canny.type()); img_fill.copyTo(img_roi); cv::Point pts[4] = { cv::Point(3 * cols_kerl, 6 * rows_kerl), cv::Point(7 * cols_kerl, 6 * rows_kerl), cv::Point(9.5 * cols_kerl, 10 * rows_kerl), cv::Point(0.5 * cols_kerl, 10 * rows_kerl) }; cv::fillConvexPoly(img_fill, pts, 4, cv::Scalar(255)); cv::bitwise_and(img_canny, img_fill, img_roi); } int lane_roi_hulf(cv::Mat& img_gray, cv::Mat& img_hulf) { cv::Mat img_fill = cv::Mat::zeros(img_gray.size(), img_gray.type()); img_fill.copyTo(img_hulf); cv::Point pts[4] = { cv::Point(0, img_gray.rows / 2), cv::Point(img_gray.cols, img_gray.rows / 2), cv::Point(img_gray.cols, img_gray.rows), cv::Point(0, img_gray.rows), }; cv::fillConvexPoly(img_fill, pts, 4, cv::Scalar(255)); cv::bitwise_and(img_gray, img_fill, img_hulf); } std::vector<std::vector<cv::Vec4i>> line_separa(std::vector<cv::Vec4i> lines, cv::Mat& img_edges) { std::vector<std::vector<cv::Vec4i> > output(2); size_t j = 0, m = 0, n = 0; cv::Point ini, fini; double slope_thresh_min = 0.3; double slope_thresh_max = 10; std::vector<double> slopes, slope_left, slope_right; std::vector<cv::Vec4i> selected_lines, sel_line_left, sel_line_right; std::vector<cv::Vec4i> right_lines, left_lines; // Calculate the slope of all the detected lines for (auto i : lines) { ini = cv::Point(i[0], i[1]); fini = cv::Point(i[2], i[3]); // Basic algebra: slope = (y1 - y0)/(x1 - x0) double slope = (static_cast<double>(fini.y) - static_cast<double>(ini.y)) / (static_cast<double>(fini.x) - static_cast<double>(ini.x) + 0.00001); // If the slope is too horizontal, discard the line // If not, save them and their respective slope if (std::abs(slope) > slope_thresh_min && std::abs(slope) < slope_thresh_max) { slopes.push_back(slope); selected_lines.push_back(i); } } // Split the lines into right and left lines double img_center = static_cast<double>((img_edges.cols / 2)); while (j < selected_lines.size()) { ini = cv::Point(selected_lines[j][0], selected_lines[j][1]); fini = cv::Point(selected_lines[j][2], selected_lines[j][3]); // Condition to classify line as left side or right side if (slopes[j] > 0 && fini.x > img_center && ini.x > img_center) { right_lines.push_back(selected_lines[j]); //right_flag = true; } else if (slopes[j] < 0 && fini.x < img_center && ini.x < img_center) { left_lines.push_back(selected_lines[j]); //left_flag = true; } j++; } while (m < right_lines.size()) { ini = cv::Point(right_lines[m][0], right_lines[m][1]); fini = cv::Point(right_lines[m][2], right_lines[m][3]); // Basic algebra: slope = (y1 - y0)/(x1x1 - x0) double slope = (static_cast<double>(fini.y) - static_cast<double>(ini.y)) / (static_cast<double>(fini.x) - static_cast<double>(ini.x) + 0.00001); double interc = fini.y - slope * fini.x; std::cout<<"slope1: "<<slope<<", interc1: "<<interc<<std::endl; slope_right.push_back(slope); sel_line_right.push_back(right_lines[m]); m++; } while (n < left_lines.size()) { ini = cv::Point(left_lines[n][0], left_lines[n][1]); fini = cv::Point(left_lines[n][2], left_lines[n][3]); // Basic algebra: slope = (y1 - y0)/(x1x1 - x0) double slope = (static_cast<double>(fini.y) - static_cast<double>(ini.y)) / (static_cast<double>(fini.x) - static_cast<double>(ini.x) + 0.00001); double interc = fini.y - slope * fini.x; std::cout<<"slope2: "<<slope<<", interc2: "<<interc<<std::endl; // If the slope is too horizontal, discard the line // If not, save them and their respective slope if (std::abs(slope) > slope_thresh_min && std::abs(slope) < slope_thresh_max) { slope_left.push_back(slope); sel_line_left.push_back(left_lines[n]); } n++; } output[0] = right_lines; output[1] = left_lines; return output; } std::vector<cv::Point> regression(std::vector<std::vector<cv::Vec4i> > left_right_lines, cv::Mat& inputImage) { std::vector<cv::Point> output(4); cv::Point ini, fini; cv::Point ini2, fini2; cv::Vec4d right_line, left_line; std::vector<cv::Point> right_pts, left_pts; cv::Point right_b; // Members of both line equations of the lane boundaries: double right_m; // y = m*x + b cv::Point left_b; // double left_m; // // If right lines are being detected, fit a line using all the init and final points of the lines for (auto i : left_right_lines[0]) { ini = cv::Point(i[0], i[1]); fini = cv::Point(i[2], i[3]); right_pts.push_back(ini); right_pts.push_back(fini); } if (right_pts.size() > 0) { // The right line is formed here cv::fitLine(right_pts, right_line, CV_DIST_L2, 0, 0.01, 0.01); right_m = right_line[1] / right_line[0]; right_b = cv::Point(right_line[2], right_line[3]); } // If left lines are being detected, fit a line using all the init and final points of the lin for (auto j : left_right_lines[1]) { ini2 = cv::Point(j[0], j[1]); fini2 = cv::Point(j[2], j[3]); left_pts.push_back(ini2); left_pts.push_back(fini2); } if (left_pts.size() > 0) { // The left line is formed here cv::fitLine(left_pts, left_line, CV_DIST_L2, 0, 0.01, 0.01); left_m = left_line[1] / left_line[0]; left_b = cv::Point(left_line[2], left_line[3]); } // One the slope and offset points have been obtained, apply the line equation to obtain the line points int ini_y = inputImage.rows; int fin_y = 470; double right_ini_x = ((ini_y - right_b.y) / right_m) + right_b.x; double right_fin_x = ((fin_y - right_b.y) / right_m) + right_b.x; double left_ini_x = ((ini_y - left_b.y) / left_m) + left_b.x; double left_fin_x = ((fin_y - left_b.y) / left_m) + left_b.x; output[0] = cv::Point(right_ini_x, ini_y); output[1] = cv::Point(right_fin_x, fin_y); output[2] = cv::Point(left_ini_x, ini_y); output[3] = cv::Point(left_fin_x, fin_y); return output; } int find_contour(cv::Mat& img_roi, cv::Mat& img_contour) { //寻找轮廓 vector<vector<Point>> contours; findContours(img_roi, contours, RETR_EXTERNAL, CHAIN_APPROX_SIMPLE); for (int i = 0; i < contours.size(); i++) { //计算面积与周长 double length = arcLength(contours[i], true); double area = contourArea(contours[i]); //外部矩形边界 Rect rect = boundingRect(contours[i]); int h = img_roi.rows - 50; //轮廓分析： if (length < 50.0 || area < 10.0) { continue; } //cout << "length:" << length << endl; //cout << "area:" << area << endl; /* //最小包围矩形 RotatedRect rrt = minAreaRect(contours[i]); //关于角度问题：https://blog.csdn.net/weixin_41887615/article/details/91411086 double angle = abs(rrt.angle); //angle < 50.0 || angle>89.0 if (angle < 20.0 || angle > 89.0) { continue; } */ if (contours[i].size() > 5) { //用椭圆拟合 RotatedRect errt = fitEllipse(contours[i]); if ((errt.angle < 5.0) || (errt.angle > 160.0)) { //if (80.0 < errt.angle && errt.angle < 100.0) { continue; //} } } drawContours(img_contour, contours, i, Scalar(255), 1, 8); } return 0; } //g++ -std=c++14 lane_test3.cpp -o lane_test3 -lopencv_core -lopencv_highgui -lopencv_videoio -lopencv_imgproc -lopencv_imgcodecs