Joint denoising and lossless/lossy compression of SAR images

Abstract

It is known that the performance of the Discrete Wavelet Transform (DWT) decreases when attempting to compress noisy images, and that it can be improved if a suitable denoising algorithm is used as a preprocessing stage. In this paper we show that this cannot be straightforwardly applied to the Integer Wavelet Transform (IWT) due to the non-linear operations involved. We propose an algorithm based on the IWT, which is capable of achieving performance nearly equal to that of the DWT, and can be satisfactorily employed for joint denoising and lossless/lossy compression. Simulation results are given in the case of SAR images.