理论
解释:在二阶导数的时候,最大变化处的值为零即边缘是零值。通过二阶
导数计算,依据此理论我们可以计算图像二阶导数,提取边缘。
l二阶导数我不会,别担心 ->拉普拉斯算子(Laplance operator)
lOpencv已经提供了相关API - cv::Laplance
处理流程:
l高斯模糊 – 去噪声GaussianBlur()
l转换为灰度图像cvtColor()
l拉普拉斯 – 二阶导数计算Laplacian()
l取绝对值convertScaleAbs()
l显示结果
API使用cv::Laplacian:
Laplacian(
InputArray src,
OutputArray dst,
int depth, //深度CV_16S
int kisze, // 3
double scale = 1,
double delta =0.0,
int borderType = 4
)
#include <opencv2/opencv.hpp>
#include <iostream>
using namespace std;
using namespace cv;
Mat src, dst, gray_src;
int main(int agrc, char** agrv) {
src = imread("C:/Users/liyangxian/Desktop/bjl/nm3.jpg");
if (!src.data) {
printf("no load..\n");
return -1;
}
const char* input_win = "input";
const char* out_put = "out_put";
namedWindow(input_win, CV_WINDOW_AUTOSIZE);
namedWindow(out_put, CV_WINDOW_AUTOSIZE);
imshow(input_win, src);
GaussianBlur(src, dst, Size(3, 3), 0, 0);
cvtColor(dst, gray_src, CV_BGR2GRAY);
Mat edge_image;
Laplacian(gray_src, edge_image, CV_16S, 3);
convertScaleAbs(edge_image, edge_image);
threshold(edge_image, edge_image, 0, 255, THRESH_OTSU | THRESH_BINARY);
imshow(out_put, edge_image);
waitKey(0);
return 0;
}