1、目的

把图片转换为灰度图，之后进行2次1倍图像尺寸缩写 ，再进行边沿检测。

2、代码

#include <opencv2/opencv.hpp>

void help(char** argv ) {

std::cout << "\n"

<< "\nExample 2-8. Combining the pyramid down operator (twice) and the Canny"

    << "\n            subroutine in a simple image pipeline"

    << "\nCall:\n" 

<< argv[0] <<" <path/image>\n"

<< "For example:\n"

<< argv[0] << " ../fruits.jpg\n"

<< std::endl;

}

int main( int argc, char** argv ) {

if (argc != 2) {

help(argv);

return 0;

}

  cv::Mat img_rgb, img_gry, img_cny, img_pyr, img_pyr2;

  cv::namedWindow( "Example Gray", cv::WINDOW_AUTOSIZE );

  cv::namedWindow( "Example Canny", cv::WINDOW_AUTOSIZE );

  img_rgb = cv::imread( argv[1] );

  cv::cvtColor( img_rgb, img_gry, cv::COLOR_BGR2GRAY);//转换为灰度图

  cv::pyrDown( img_gry, img_pyr ); //缩写一倍尺寸

  cv::pyrDown( img_pyr, img_pyr2 );//缩写一倍尺寸

  cv::Canny( img_pyr2, img_cny, 10, 100, 3, true ); //边沿检测

  cv::imshow( "Example Gray", img_gry );

  cv::imshow( "Example Canny", img_cny );

  cv::waitKey(0);

}

3、结果