Novel subband DCT image coding

In transform image coding, an image is usually divided into subimages before processing. Though a smaller size subimage is efficient for computation, it introduces blocking effect. However, an increase in subimage size results in an increase in both storage and computational requirements. This paper proposes an efficient coding approach to simulate a full-size DCT image coding with subband decomposition technique in such a way that only a small increase in storage is required. Compared with the conventional DCT transform coding, this coding technique has the advantages of (i) no blocking effect, (ii) higher quality reconstructed images and (iii) less computational effort required.