16-bit range representation

I have a simple question (I hope). I am analyzing TIF images from a 16-bit scanner. The scanner reports these as numbers from 1 - 65536, while CellProfiler reports numbers as fractions from 0 - 1.0. To convert CellProfiler’s fractions back to the original format, is it just a matter of multiplying by 65536? That is, 0.5 = 32768, 0.25 = 16384? This seems implicit in some of the discussion on rescaling 12-bit to 16-bit images but I cannot find it actually stated anywhere.


Hi Earl,

LoadImages records the native bit-depth of the input image as a measurement; it then divides by this scaling factor to convert the image to range from 0 to 1. If you need to use this factor in a pipeline for a calculation, you can select it as a measurement operand in ImageMath or CalculateMath. It’s also written to the per-image spreadsheet if you need to make use of it there.

However, keep in mind that the identification modules expect a rescaled image as input. So if you want to convert the images back to their native range, you should only apply the measurement modules to these rescaled images.