# Normalized image for shape from shading

Im going over this paper by Vogel A Direct Numerical Approach to Perspective Shape-from-Shading. I have a problem in the way they define the normalized image brightness I(x) as:

I(x) = E(x)/sigma

where sigma "depends on the albedo of the surface and the intensity of the light source".

I am estimating the albedo with the formula given here. As the albedo is in the range of [0, 1], and the max light intensity is 255, I'm guessing that sigma must be atmost 255. However, in the paper the authors have taken a value of 1000. Later for another example of Mozart image, the authors have assume a value of 4*10^4.

It would be great if someone could explain how they arrive at these values and why, and where I'm wrong.

Thanks

• Where are you getting the max light intensity of 255? That sounds like a limitation of some system that uses 8-bit values for lights. In general, light intensities can be any positive value. Lights span a huge range of magnitudes in brightness, e.g. from candle flames up to the sun. – Nathan Reed Jan 29 at 0:03