In my previous article, I compared two images of the same size pixel by pixel. To further enhance its functionality, I now search for a small image (I’m going to call it a jigsaw) in the big image pixel by pixel and return the coordinates of the top left corner pixel location.
I cropped a 36x36 image from a.png and named it c.png
After that I edited c.png and named it d.png so that the jigsaw could not be found anymore.
It takes ~11 seconds to search for a 36x36 pixel jigsaw in a 360x360 image. Next I convert all the images into grayscale and name it ag.png, cg.png and dg.png
convert a.png -type grayscale ag.png
Now lets retest the search with the grayscale images.
It seems that it is faster with grayscale images, it took about ~4 seconds to search.
Related articles
Image Compare By Pixels
Needle In Haystack Search (version 2)
