An Improved Neigh Shrink in Hybrid Wavelet Transform for Image Compression

Abstract

Rapid progress in mobile communication, demands high-speed multimedia data processing. This has lead the compression technology to speed up the process by reducing the size of data without disturbing image quality. This paper attempts to develop the quality of compression by using improved neigh shrink in hybrid wavelet algorithm to compress an image. After wavelet decomposition, one-dimensional Discrete Cosine Transform is applied to decorrelate approximate coefficients and stored as T-matrix. The detail coefficients are thresholded using ‘improved neigh shrink' and Eliminate Zero and Store Data algorithm are applied to eliminate redundancy and stored as reduced array. The compressed approximate and detailed coefficients are encoded by arithmetic coding. The simulated results show that proposed algorithm has significant improvement in image quality in terms of PSNR and SSIM when compared with existed wavelet-based compression methods including JPEG 2000 at high compression rate. Keywords: Discrete Wavelet Transform (DWT), T-matrix coding, improved Neigh Shrink, EZSD.