Introduction - If you have any usage issues, please Google them yourself
Traditional edge detection algorithm can only target the pixel level, you can not get an accurate result, the pixel sub-pixel is further broken down, as a result of more accurate