Phe Research on Flatness Pattern Recognition Based On Improved Direct GA-BP Neural Network

摘要

In this paper, an improved direct recognitiong approach to flatness pattern with BP neural network Optimized by genetic algorithm (GA) is proposed.lt can not only efficiently sovle the problem of different topologic configurations with different widths in the traditional shape flatness pattern recognition neural network, but also improve practicability and precision. Compared to shape pattern recognition using BP network based on fuzzy distance(FBP),The simulation results show that the new method is a feasible methord to recognize flatness pattern.