SAR Image Speckle Noise Suppression Based on DFB Hidden Markov Models Using Immune Clonal Selection Thresholding

Synthetic aperture radar (SAR) images are inherently affected by multiplicative speckle noise, which is due to the coherent nature of the scattering phenomenon. This paper proposes a novel DFB-based algorithm with hidden Markov modeling, which reduces speckle in SAR images while preserving the structural features and textural information of the scene, and introduces evolutionary computation theory - immune clonal selection (ICS) method to optimize threshold avoiding the drawback of experiential threshold. We compare our proposed method to wavelets techniques applied on real SAR imagery and we quantify the achieved performance improvement.