Scalable image compression using combined wavelet transform and vector quantization

Spatial scalability is a feature referring to image representation in different sizes which is particularly useful in a multiuser environment with varying resolution requirements, as well as in image browsing applications. In this paper we propose a scalable image compression scheme which exploits the spatial scalability of wavelet transform based coding techniques and the high compression performance of vector quantization (VQ). Although VQ is a powerful technique for low bit rate image compression, the label stream in VQ is not scalable. We first propose a scalable VQ algorithm (SVQ) where the label stream is made scalable such that a smaller-size image can be obtained by decoding a portion of the bitstream. This image can be further enhanced in size by progressively decoding the remaining bits. To ensure partial decodability of VQ labels separate codebooks are required one for each spatial resolution. We then propose a combination of wavelet transform and SVQ technique (WSVQ) for spatial scalability. Here, the input image is decomposed into a three level wavelet decomposition where the resulting ten sub-images are reordered to form four spatial resolution images. The wavelet coefficients are then vector quantized using the SVQ scheme and a non-linear interpolative vector quantization (NTVQ) technique. Simulation results confirm the substantial reductions in bit rate and superior subjective image quality at each spatial resolution using the proposed algorithm, at a significantly reduced computational complexity.