ENHANCEMENT OF COMPRESSED IMAGES BY OPTIMAL SHIFT-INVARIANT WAVELET PACKET BASIS

A novel postprocessing method based on the optimal shift-invariant wav elet packet (SIWP) representation and wavelet shrinkage is proposed to enhance compressed images. At the encoder, the optimal (in the mean s quare error sense) SIWP basis is searched using a fast optimization al gorithm and the location of the best basis in the entire SIWP library is transmitted as overhead information to the decoder. The selected ba sis is jointly optimal in terms of both the time-frequency tiling and the relative time-domain offset (or shift) between a signal and its wa velet packet representation. After the decoder reconstructs the compre ssed image, the postprocessor performs wavelet shrinkage using the opt imal basis. Due to its powerful adaptability, the method is shown to a chieve a better trade-off between enhancement performance and decoder complexity than both the orthonormal wavelet transform and the undecim ated wavelet transform-based methods. (C) 1998 Academic Press.