I'm trying to take a raw 16 bit image and convert it to a 32 bit float, normalized from 0.0 to 1.0
If the image header states that it has a depth of 16 and 1 channel then does GREY describe that its 1 or 2 bytes / per sample?
What role does the destination image structure have on the conversion process? Is expected to be the preconstructed?
P.S. Do people here talk C or just Python?