Quality enhancement of adaptively compressed images using bit plane removal

Abstract

We propose a quality enhancement algorithm for linear adaptive compression using bit plane reduction. Compression is based on the observation that least significant bit(LSB) planes can be removed without affecting the perceptual quality of the image. It also makes use of the fact that the bits in a bit plane repeat and hence can be compressed losslessly by encoding. At the receiver end, the quality of decompressed image is improved by random bit replacement. The proposed method can be used for image quality enhancement after lossy compression or near lossless compression. The proposed method has comparable compression ratio at better PSNR and less computational complexity in comparison to JPEG 2000. It achieves better compression ratio in comparison to TIFF at same perceptual quality. The time complexity of the proposed method is only O(n).