These functions find the row and column gradients of an image, M, resulting in a horizontally or vertically selective edge detection. Edge detection is used to enhance image readability for certain types of features that depend on boundaries. Gradients also result in a visual embossing effect.

These two functions create row and column differences in an image, M, by subtracting successive rows (or columns) from each other. In each case the resulting matrix has one undefined row or column. This is set equal to zero.