Cannot convert PIL image from Paint.NET to numpy array

Question

I am getting a ValueError in numpy when performing operations on images. The problem seems to be that the images edited by Paint.NET are missing the RGB dimension when opened using PIL and converted to a numpy array.

Answer 1

If PIL is giving you a 861x1091 image when you are expecting an 861x1091x3 image, that is almost certainly because it is a palette image - see here for explanation.

The simplest thing to do, if you want a 3-channel RGB image rather than a single channel palette image is to convert it to RGB when you open it:

im = Image.open(path).convert('RGB')