ON STUDY OF FIIVITE•STATE VECTOR QUANTIZATION FOR IMAGE COMPRESSION

Abstract

In image data compression an image is represented by as few bits as possible while maintaining acceptable subjective quality of picture. Using Finite-State Vector Quantization (FSVQ) for image data compression, good subjective quality of image can be obtained at very low bit rate. For reducing degradations in block boundaries and encoding complexity, the above technique is subdivided in SMVQ and OMVQ. It has been observed from the simulation results that good subjective as well as quantitative quality of an image can be obtained. PSNR and MSE have been used as parameters for measure of quality of images.