Enhanced AMBTC for Image Compression using Block Classification and Interpolation

Abstract

ABSTRACT Block Truncation Coding (BTC) is a fast moment preserving lossy image compression technique. This paper proposes a original image. Though some data is removed, for Human two-level compression scheme. In the first level of compression, the blocks are categorized into high-detail blocks and low-detail blocks. The high detail blocks are stored or transmitted in the form of a bit plane (16 bits), and two statistical moments; high mean and low mean (16 bits). Only the mean values are stored for low detailed blocks. In second level of compression, the interpolation technique is incorporated as part of the proposed method for further bit rate reduction. The performance of the method is tested with standard images Lena, Cameraman Boats, Bridge, Baboon and Kush. The proposed scheme achieves better coding efficiency with a slight degradation in PSNR when compared to existing methods. General Terms Image Compression Keywords image compression; high-detail block; low-detail block; interpolation; bpp; PSNR;