Application of ACO-LMBP Hybrid Neural Network Algorithm in Image Denoising

2018 International Conference on Advanced Electronic Materials, Computers and Materials Engineering

材料科学;无线电电子学;计算机科学

Wang, Hai Jun^1 ; Tao, Jin^1 ; Neimule, Menke^1

Ordos Institute of Technology, Inner-Mongolia-Ordos

017000, China^1

关键词: Ant colony algorithms;    BP neural networks;    Global search ability;    Hybrid neural networks;    Image denoising algorithm;    LM-BP neural network;    Slow convergences;    Wiener filtering;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/439/3/032120/pdf
DOI  :  10.1088/1757-899X/439/3/032120

来源: IOP

【 摘 要 】

In order to overcome the disadvantages of poor global search ability, slow convergence speed and easy to fall into local minimum in the traditional BP neural network in image denoising, a hybrid ACO-LMBP neural network image denoising algorithm based on ant colony algorithm and LMBP algorithm is proposed. ACO-LMBP hybrid neural network algorithm has both the high speed of LMBP algorithm and the global nature of ACO algorithm. It can improve the problems of BP algorithm model very well. By comparing with the image denoising effect of Wiener filtering, BP, LMBP and PSO-LMBP model, the denoising model using the ACO-LMBP neural network algorithm has better denoising effect.