Genetic Programming for Image Analysis

Abstract

This paper describes an approach to using GP for image analysis based on the idea that image enhancement, feature detection and image segmentation can be re-framed as image filtering problems. GP can be used to discover efficient optimal filters which solve such problems. However, in order to make the search feasible and effective, terminal sets, function sets and fitness functions have to meet some requirements. In the paper these requirements are described and terminals, functions and fitness functions that satisfy them are proposed. Some preliminary experiments are also reported in which GP (with the mentioned characteristics) is applied to the segmentation of the brain in magnetic resonance images (an extremely difficult problem for which no simple solution is known) and compared with artificial neural nets. 1 Introduction Genetic Programming (GP) has been applied successfully to a large number of difficult problems like automatic design, pattern recognition, robotic control, synt...