Improved Digital Image Compression using Modified Single Layer Linear Neural Networks

Abstract

Compression is solved by using Wavelet-Modified Single Layer Linear Forward Only Counter Propagation Network (MSLLFOCPN) technique. Form the wavelets it inherits the properties of localizing the global spatial and frequency correlation from wavelets. Function approximation and prediction are obtained from neural networks. As a result counter propagation network was considered for its superior performance and the research enable us to propose a new neural network architecture named single layer linear counter propagation network (SLLC). The combination of wavelet and SLLC network were tested on several benchmark images and the experimental results shows that an enhancement in picture quality, compression ratio and approximation or prediction comparable to existing and traditional neural networks.