在学习opencv的时候,对一张照片,需要标注照片上物体的不规则轮廓。
如图:
使用opencv进行物体的轮廓处理,关键在于对照片的理解,前期的照片处理的越好最后调用api出来的结果就越接近理想值。
提取照片中物体分如下三步:
#include <opencv2/opencv.hpp>
#include <iostream>
#include <math.h>
using namespace cv;
using namespace std;
Mat src, dst, gray_src;
char input_image[] = "input image";
char output_image[] = "output image";
int main(int argc, char ** argv){
src = imread("case6.jpg");
if (src.empty()){
printf("colud not load image ..\n");
return -1;
}
namedWindow(input_image, CV_WINDOW_AUTOSIZE);
namedWindow(output_image, CV_WINDOW_AUTOSIZE);
imshow(input_image, src);
// 均值降噪
Mat blurImg;
GaussianBlur(src, blurImg, Size(15, 15), 0, 0);
imshow("input image", src);
// 二值化
Mat binary;
cvtColor(blurImg, gray_src, COLOR_BGR2GRAY);
threshold(gray_src, binary, 0, 255, THRESH_BINARY | THRESH_TRIANGLE);
imshow("binary", binary);
// 闭操作进行联通物体内部
Mat morphImage;
Mat kernel = getStructuringElement(MORPH_RECT, Size(3, 3), Point(-1, -1));
morphologyEx(binary, morphImage, MORPH_CLOSE, kernel, Point(-1, -1), 2);
imshow("morphology", morphImage);
// 获取最大轮廓
vector<vector<Point>> contours;
vector<Vec4i> hireachy;
findContours(morphImage, contours, hireachy, CV_RETR_EXTERNAL, CHAIN_APPROX_SIMPLE, Point());
Mat connImage = Mat::zeros(src.size(), CV_8UC3);
for (size_t t = 0; t < contours.size(); t++){
Rect rect = boundingRect(contours[t]);
if (rect.width < src.cols / 2) continue;
if (rect.width > src.cols - 20) continue;
double area = contourArea(contours[t]);
double len = arcLength(contours[t], true);
drawContours(connImage, contours, t, Scalar(0, 0, 255), 1, 8, hireachy);
printf("area of star could : %f \n", area);
printf("lenght of star could : %f \n", len);
}
imshow(output_image, connImage);
waitKey(0);
return 0;
}
?