IMAGE COMPRESSION USING MULTILAYER NEURAL NETWORKS BASED ON FAST BACTERIAL SWARMING ALGORITHM

摘要

A novel multilayer neural network based on Fast Bacterial Swarming Algorithm (FBSA) for image compression is presented in this paper. The proposed novel neural network combines the traditional structure of two-layer symmetric Artificial Neural Network (ANN) and the newly presented FBSA, which demonstrates quick and good performance for high-dimensional function optimization problems. Comparison of the qualities of the reconstructed images had shown the effective improvement using the proposed algorithm than using the conventional BP network.