the number of bits of an image read with imread

[Eric Nowak]

Hi all,
I am using octave to read 8bits and 16bits image with imread().
I need to convert the image to floats in [0,1], e.g. dividing the image values by 2^bitnb.
The pb is, I can't figure out what the bit nb is.

Of course, I can't look for the max value of an image to figure it out because:
- it is too slow
- a 16bits image may have values under 255

Any idea how to solve the problem?

Thanks,
Eric