基于PSO—BP神经网络的图像压缩算法  

Image Compression Algorithm Based on PSO-BP Neural Networks

作　　者：王书宇[1] 施宁[1] 李子杰[1] WANG Shu-yu,SHI Ning,LI Zi-Jie （Artillery Academy of PLA, Hefei 230031, China）

机构地区：[1]解放军炮兵学院,安徽合肥230031

出　　处：《电脑知识与技术》2007年第12期1409-1411,共3页Computer Knowledge and Technology

摘　　要：本文阐述了使用BP神经网络压缩图像的方法和粒子群算法（PSO）的原理。为提高BP算法的训练速度和图像重建质量．本文设计了一种利用PSO—BP网络进行图像压缩的算法，该算法结合了PSO算法和BP算法的优点，将BP网络的训练过程分为两个阶段。实验表明，利用该算法压缩图像，不仅速度较快，而且重建后的图像质量有明显提高。This paper introduces the technique of the BP artificial neural network in the application of computer image compression and the principle of Particle Swarm Optimizer （PSO） algorithm. In order to enhance the training speed of the BP algorithm, the quality of image compression, this paper design a algorithm based on BP-PSO network for image compression and decompression, the algorithm consider the merits of the PSO and BP algorithms, the training process of BP network is divided into two steps. The experiments indicate that the convergence speed and precision of BP networks using the improved algorithm are enhanced greatly. The compression ratio and reconstruction quality of images are improved distinctly.

关 键 词：BP网络 图像压缩 PSO算法 

分 类 号：TP183[自动化与计算机技术—控制科学与工程;自动化与计算机技术—控制理论与控制工程]