Image Compression and Reconstruction Using a New Approach by Artificial Neural Network

In this paper a neural network based image compression method is presented. Neural networksoffer the potential for providing a novel solution to the problem of data compression by its abilityto generate an internal data representation. This network, which is an application of backpropagation network, accepts a large amount of image data, compresses it for storage ortransmission, and subsequently restores it when desired. A new approach for reducing trainingtime by reconstructing representative vectors has also been proposed. Performance of thenetwork has been evaluated using some standard real world images. It is shown that thedevelopment architecture and training algorithm provide high compression ratio and low distortionwhile maintaining the ability to generalize and is very robust as well.