Context adaptive image coding using lapped transform

The efficient embedded image coding algorithm is proposed. This is the implement of embedded wavelet image coding technique to image which is transformed with lapped transform - block transform without blocking effect. With appropriate coefficients rearrangement, the lapped transform coefficients can be encoded efficiently with algorithm designed for wavelet coefficients. The context adaptive wavelet difference reduction (CAWDR), which is competitive to other state-of-the-art coder, SPIHT and JPEG2000, is wavelet coder algorithm applied in this paper. The binary output bit-stream is further encoded with well-organized multiple context adaptive arithmetic coder. The coder preserves embedded property and exact rate control, and has competitive rate-distortion result to other coders in the literature. The PSNR value of the proposed coder can reach 1.3 dB higher than that of image reconstructed by JPEG2000 at the same compression ratio.