Enhancement and FPGA implementation of ANFIS algorithm in digital image processing

摘要

Adaptive Neuro Fuzzy Inference System (ANFIS) is a kind of neuro-fuzzy model, combining fuzzy system and neural network. It incorporates the human-like reasoning style of fuzzy systems through the use of fuzzy sets and a linguistic model consisting of a set of IF-THEN fuzzy rules. Based on Takagi-Sugeno Fuzzy Inference System, ANFIS focus on the accuracy and it is widely used in control and identification systems. However, when the fuzzy rule base is large, it proved to be slow because of the computation time. This paper introduces an enhanced algorithm, implemented on FPGA platform, to speed up the standard ANFIS algorithm in digital image processing.