Embedded image compression using DCT based subband decomposition and SLCCA data organization

Abstract

Wavelet transform provides harmonic space-frequency localization and great energy compaction, but with generally high computational complexity. In this paper, an 8/spl times/8 fast discrete cosine transform (DCT) approach is adopted to perform subband decomposition, followed by SLCCA data organization and entropy coding. Simulation results showed that the embedded DCT-SLCCA image compression reduced the computational complexity to only a quarter of the wavelet based subband decomposition while the average peak signal-to-noise ratio (PSNR) degraded at the same bit rate by only 0.6 dB. The subjective image quality degradation was almost unnoticeable. Moreover, due to 8/spl times/8 fast DCT hardware implementation being commercially available, the proposed DCT-SLCCA has the potential for high performance high speed image coding and transmission.