文章基本信息

标题：A Fast Compression Technique by Improving the Parameters and Structure of Neural Network Using Improved GA and Switches Structure
本地全文：下载
作者：Digvijay Singh Thakur ; Prabhakar Sharma
期刊名称：International Journal of Innovative Research in Science, Engineering and Technology
印刷版ISSN：2347-6710
电子版ISSN：2319-8753
出版年度：2017
卷号：6
期号：10
页码：19855
DOI：10.15680/IJIRSET.2017.0610123
出版社：S&S Publications
摘要：Genetic algorithm and neural network based techniques are very powerful method for solving real lifeproblems and has been widely studied for its applications in areas of image compression. For image compression.GAbased learning for neural network is suffering from curse of very slow convergence and poor quality of compression.Neural networks for tuning usually have a fixed structure. The number of connections has to be large enough to fit agiven application. This may cause the neural network structure to be unnecessarily complex and increase theimplementation cost. The current work proposes a neural network with switches. By doing this, the proposed neuralnetwork can learn both the input-output relationships of an application and the network structure. The experimentalresults show that the technique performed efficiently then the algorithm version without switches.
关键词：Image Processing; Image Compression; Genetic Algorithm; Switches.