The purpose of this paper is to show how one can use entropies of deterministic functions (as previously defined by the author) in order to analyze some questions related to machine vision. The main advantage of this approach is that it provides information theoretic methods for solving problems which basically do not refer to probability distributions. After a short qualitative background on deterministic functions, one applies this theory to edge finding, image segmentation, transfer of information defined by brightness functions, and image processing. Some more theoretical details on the entropies of deterministic functions are given in the appendix.
