Image compression using non-stationary and inhomogeneous multiresolution analyses

New methods for lossy image compression based on wavelet theory are introduced. Unlike the classical wavelet decomposition scheme it is possible to have diierent scaling and wavelet lters at every scale by using non-stationary multiresolution analyses. For the bidimensional case inhomogeneous multires-olution analyses using diierent wavelet lters for the two variables are introduced. Beyond it, these two methods are combined. All this freedom is used for compact image coding. The idea is to build out of the lters in a library that special non-stationary and/or inhomogeneous multiresolution analysis, that is best suited for a given image in the context of compact coding (in the sense of optimizing certain cost-functions). In order to identify the optimal ltercombination, an adaptive algorithm of low complexity is developped. The usefulness of these new schemes for image compression is demonstrated.