Computing entropy maps of finite-automaton-encoded binary images

Abstract

Finite automata are being used to encode images. Applications of this technique include image compression, and extraction of self similarity information and Hausdorff dimension of the encoded image. J Jürgensen and Staiger [7] proposed a method by which the local Hausdorff dimension of the encoded image could be effectively computed. This paper describes the first implementation of this procedure and presents some experimental results showing local entropy maps computed from images represented by finite automata.