Comparison of intelligent classification techniques applied to marble classification

Automatic marbles classification based on their visual appearance is an important industrial issue. However, there is no definitive solution to the problem, mainly due to the presence of randomly distributed high number of different colors and due to the subjective evaluation made by human experts. In this paper, we present a study of soft computing classification algorithms, which proved to be a valuable tool to be applied in this type of problems. Fuzzy, neural, simulated annealing, genetic and combinations of these approaches are compared. Color and vein classification of marbles are compared. The combination of fuzzy classifiers optimized by genetic algorithms revealed to be the best classifier for this application.