PROJECT TITLE :

Despeckling of SAR Image Using Generalized Guided Filter With Bayesian Nonlocal Means

ABSTRACT:

Because of the coherent nature of the scattering phenomenon, artificial aperture radar (SAR) images are inherently contaminated by multiplicative speckle noise, that makes despeckling continuously a elementary problem for SAR image processing. Motivated by the thought of the guided image filter, we have a tendency to propose an extended despeckling theme named generalized guided filter with Bayesian nonlocal means (GGF-BNLM). Our main contributions are as follows: one) We successfully deduce the nonlinear weight kernel of the GGF-BNLM framework; and a couple of) we have a tendency to construct the steerage image using homogeneity analysis of local regions and the utmost-probability rule. Visual and quantitative experiments conducted on artificial speckle pictures and real SAR pictures show that our method notably suppresses speckle with unperceivable detail blurring and better preserving of point-kind strong scatters, which outperforms many classical and state-of-the-art strategies.