Modifying integer wavelet transforms for scalable near-lossless image coding

In near-lossless image coding, each reconstructed pixel of the decoded image differs from the corresponding one in the original image by not more than a pre-specified value δ. Such schemes are mainly based on predictive coding techniques, which are not capable of scalable decoding. Lossless image coding with scalable decoding is mainly based on integer wavelet transforms. In this paper, methods to modify integer wavelet transforms for near-lossless image coding with scalable decoding features are presented. This is achieved by incorporating the near-lossless quantisation process, driven by δ, into lifting steps (online quantisation). Two online quantisation techniques based on 1-D and 2-D transforms are presented. They outperform the pre-quantisation based near-lossless image coding method in both bit rate and rms error performances. Further, they result in both subjectively and objectively superior performance in spatial and bit rate scalable decoding. The 2-D online scheme results in comparable performance with JPEG-LS, which is not capable of scalable decoding. It is evident from this research that with these novel schemes, scalable decoding features can be integrated into near-lossless coding with only a small increase in bit rate compared to those achieved in JPEG-LS.