Hi All, I am new to python and image processing in general. I am having trouble working with 12 bit images to do some very simple calculations. Here is some example code:

#!/usr/local/bin/ipython

#import Imaging Library

from PIL import Image

from skimage import io, filters

from skimage import img_as_float

from skimage import exposure

#import pylab for plotting

from pylab import *

Aim = io.imread("A.tiff")

Bim = io.imread("B.tiff")

Cim = io.imread("C.tiff")

Dim = Aim - Cim

Eim = Bim - Cim

#print min and max values of the background subtracted images

print("min %d max %d" % (Aim.min(),Aim.max()))

print("min %d max %d" % (Bim.min(),Bim.max()))

print("min %d max %d" % (Dim.min(),Dim.max()))

print("min %d max %d" % (Eim.min(),Eim.max()))

Input images A,B and C are 12 bit greyscale TIFFs. Output:

min 0 max 4095

min 0 max 4095

min 0 max 65533

min 0 max 65533

The input image data response to min and max make good 12bit sense, but it is totally beyond me how I am getting 16bit responses for D and E. I want to understand what's happening here so that I don't get bitten when I try to do transformations that require me to use floating point math.

thanks, Johnny