Color consistency classification system based on fuzzy neural network

Abstract

Tile color consistency inspection and classification is the most difficult problem in tile online surface quality inspection. This paper constructed a fuzzy neural network which combined artificial neural network and fuzzy technique and solved this problem. Artificial neural network was applied to solve the problem that there is no certain rule because the type of tile is various and the result needs accord to the ones of human vision. Fuzzy technique which can simulate abstract thinking was applied in the input and output of artificial neural network to offset the shortage of the method that only applied artificial neural network which can only simulate virsulizying thinking. The result showed that this method performed better in speed and precision than traditional method or only using neural network techniques.