The method aims to emphasize local differences between images and is not particularly designed for this. But maybe it is possible to get image B from image A +

The method aims to emphasize local differences between images and is not particularly designed for this. But maybe it is possible to get image B from image A + LDM. I need to think on this point. Color : next after being able to deal with gray-level images (soon).