Evolving fuzzy classifier system using PSO for RoboCup vision applications

In this paper we propose a color classification algorithm in which an evolutionary design optimizes a fuzzy system for color classification and image segmentation. This system works with the least number of rules and has minimum error rate by the mean of particle swarm optimization (PSO) method. In this approach each particle of the swarm codes a set of fuzzy rules. During evolution, each member of a population tries to maximize a fitness criterion which has designed to raise classification rate and to reduce the number of rules. Finally, the particle with the highest fitness value is selected as the best set of fuzzy rules for image segmentation. Fuzzy sets are defined on the H, S and L components of the HSL Color Space to provide a fuzzy logic model which aims to follow the human intuition of Color Classification. Color-based vision applications face the challenge of color variations by illumination. The final system designed by this method is adaptive to continuous variable lighting according to its evolving-fuzzy nature. In this method parameters setting is done only once .The experimental results in RoboCup leagues demonstrate that the presented approach can be very robust to noise and light variations.