Wavelet time-scale dependencies for signal and image compression

In this paper, an evolution law through scales for modulus maxima of the wavelet transform of a signal is proposed. It corresponds to an atomic approximation of wavelet detail bands. Although this result is very general and useful for various fields of image processing, we focus on its application to signal and image compression. We propose a simple trick for taking advantage from the law without directly solving its corresponding partial differential equation. Preliminary experimental results show that the proposed approach outperforms available compression techniques.