BP neural network image restoration algorithm based on self-adaption genetic algorithm

Abstract

The invention relates to a BP neural network image restoration algorithm and especially relates to the BP neural network image restoration algorithm based on a self-adaption genetic algorithm. The algorithm comprises the following steps: S1) population initialization; S2) design of a fitness function; S3) selection design; S4) cross-over design; S5) variation design; and S6) crossing-over rate and variation rate dynamic adjustment. The algorithm has smaller prediction error and more reliable stability, and enhances algorithm robustness.

Description

technical field [0001] The invention relates to a BP neural network image restoration algorithm, in particular to a BP neural network image restoration algorithm based on an adaptive genetic algorithm. Background technique [0002] Since the 1960s, image restoration technology has developed rapidly, and classic restoration methods such as inverse filtering, Wiener filtering, and least square filtering have appeared. However, these traditional restoration methods are difficult to grasp the changes of many point spread functions (PSF), so it is impossible to obtain an accurate restoration model. BP neural network can solve this problem, and its nonlinear mapping ability can realize the approximation to multidimensional functions. [0003] However, when the BP neural network is used for image restoration, the error surface of the BP algorithm usually has multiple extreme points, so the algorithm is easy to fall into the local optimal solution. And when the BP neural network s...

AI Technical Summary

Problems solved by technology

Excessively high crossover rate may destroy the model of better individuals, and too small crossover rate will delay the generation of new individuals, leading to premature algorithm and stagnation

[0013] The above-mentioned traditional genetic algorithm tends to fall into local convergence and cause stagnation

Images