Bayesian Approach to Wavelet-based Speckle Filtering of SAR Images
YU Wen-xian · Acta Simulata Systematica Sinica · 2004
We propose a new and efficient speckle filtering algorithm in the wavelet domain for SAR images. Although a wavelet transform has decorrelating properties, structures in images, like edges, are never decorrelated completely, and these structures appear in the wavelet coefficients. We therefore introduce a geometrical prior model, and combine this with the conditional probability into a Bayesian framework. In this way, we can compute for each coefficient the probability of being 搒ufficiently clean. The manipulation of the wavelet coefficients is consequently based on the obtained probability. The main novelty is the use of realistic distributions of the wavelet coefficients which represent mainly speckle noise on the one hand and those that represent the useful signal on the other. We propose analytic models for these distributions, and the automatic computation of their parameters directly from a given SAR images. Finally, we compare our technique with soft and hard thresholding methods applied on SAR images and quantify the achieved performance improvement.