Performance of optimized fuzzy edge detectors using particle swarm algorithm

Abstract

The purpose of the paper is to compare the performance of various fuzzy edge detectors which have been optimized by Particle Swarm Optimization (PSO). Three different type edge detectors Classical fuzzy Heuristic (CFH), Gaussian rule based (GRBF) and Robust Fuzzy Complement (RFC) are used. These edge detectors are effective in detecting edges, however the edges are thick. This paper proposes the used of particle swarm optimization algorithm as a method of producing thin and measurable edges. The fuzzy edge detectors are used in the initial swarm population and the objective function. The performance is based on the consistency of the visual appearance, fuzzy membership threshold and the number of complete edges detected. All three optimized edge detector performs reasonably well but CFHPSO outperform the rest. © 2010 Springer-Verlag.