Improved colour decorrelation for lossless colour image compression using the LAR codec

Next generations of still image codecs should not only have to be efficient in terms of compression ratio, but also propose other functionalities such as scalability, lossy and lossless capabilities, region-of-interest coding, etc. In previous works, we have proposed a scalable compression method called LAR, for Locally Adaptive Resolution, that covers these requirements. In particular, the Interleaved S+P scheme offers an efficient mean to compress images. In this paper, three modifications of this coder are proposed to extend its capabilities to the lossless coding of colour images. Firstly, decorrelation of the image components is introduced by using reversible colour transforms. Secondly, an adaptive decorrelation of the components is introduced. Finally, a classification between the image components is introduced. Results are then discussed and compared to the state of the art, thus revealing high compression performances of our coding solution.