Generalized SAR Despeckling Based on DTCWT Exploiting Interscale and Intrascale Dependences

Abstract

In this letter, dual tree complex wavelet transform (DTCWT)-based despeckling algorithm is proposed for synthetic aperture radar (SAR) images, using multivariate statistical theory. The DTCWT coefficients in each subband are modeled with a multivariate Cauchy probability density function (pdf) which takes into account the statistical dependency between the wavelet coefficients, their neighbors and coefficients across scales. Generalized expressions are derived for the dispersion parameter in the multivariate Cauchy pdf using the fractional moments and for the multivariate maximum a posteriori estimator. Experimental results show that the proposed method based on multivariate Cauchy prior achieves better performance in terms of equivalent number of looks, peak signal-to-noise ratio, and mean structural similarity index matrix.