Sofm And Vector Quantization For Image Compression By Component: Review

This paper present artificial neural network using SOFM and vector quantization (VQ) which has gained many research concentration and importance to improve the image quality after lossless comfpression to reduce image size. The aim of this research is to investigate image compressing using SOFM 2D K-Map with vector quantization methods and arithmetic coding of lossless compression methods, SOFM and VQ are adopted technique to improve the image compression effective. However, this paper proposed and investigated SOFM based on VQ by components, the proposed new approach which provide pre-processing for subsampling into high-pass filter and low-pass filter, low pass-filter subsampling goes immediately to lossless compressing for entropy coding and as presented here is used arithmetic coding, and high-pass filter to vectorization image block then to vector quantization using SOFM. The investigation shows there are many methods used types of filters as first stage of pre-processing in image compression before entropy.