Hands-On GPU-Accelerated Computer Vision with OpenCV and CUDA

Scharr filters

As Sobel does not provide rotation symmetry, a Scharr operator is used to overcome that by using different filter masks as follows:

As can be seen from the mask, the Scharr operator gives more weight to central rows or central columns to find edges. The program to implement a Scharr filter is as follows:

#include <iostream>#include "opencv2/opencv.hpp"int main (){  cv::Mat h_img1 = cv::imread("images/blobs.png",0);  cv::cuda::GpuMat d_img1,d_resultx,d_resulty,d_resultxy;  d_img1.upload(h_img1);  cv::Ptr<cv::cuda::Filter> filterx,filtery;  filterx = cv::cuda::createScharrFilter(CV_8UC1,CV_8UC1,1,0); filterx->apply(d_img1, d_resultx); ...