Low Complexity Embedded Image Compression Algorithm using Subband Coding of the DCT Coefficients

Abstract

This paper proposes a new algorithm termed Quad Tree-Discrete Cosine Transform (QT-DCT). QT-DCT is a low complexity algorithm that combines the good energy compaction ability and the simplicity of the DCT used by the Joint Photographic Expert Group (JPEG) image compression standard, and the global and embedding properties of Subband Coding (SBC) systems. The proposed QT-DCT reorders the DCT coefficients into subbands. Then these subbands are coded using quad-tree decomposition method. The main contribution of the proposed QT-DCT algorithm is that it has better PSNR and lower complexity than JPEG image compression standard and some existing algorithms that have about the same complexity. In addition, QT-DCT produces an embedded compressed bit-stream which is able to achieve any compression bit-rate very easily. This is a very desirable feature for the modern requirements represented by the Internet.