Simulation of fringe order of photoelastic samples based on Neural Network

The purpose of this paper is to solve the problem of traditional photoelastic fringe recognition and automatic grading through the BP artificial neural network. The nonlinear mapping relationship between the RGB value of each point on the photoelastic picture and the fringe order of the corresponding point is verified to resolve the problem. The simulation analysis is carried out by the BP neural network. The simulation analysis data shows that it has a favorable effect on fringes below grade 4.