Next generations of still image codecs should have not only to be efficient in terms of compression ratio, but have also to propose other functionalities such as scalability, lossy and lossless abilities, region of interest coding, etc. In previous works, we have proposed the LAR compression method covering these requirements. In particular, the RWHaT + P pyramid has recently been presented as a powerful reversible scalable coding technique. This paper introduces new significant improvements by the use of an inter-coefficient classification method. Results are discussed and compared to the state of the art.
展开▼
