Despeckling of SAR Images by Optimizing Averaged Power Spectral Value in Curvelet Domain
S. Md. Mansoor Roomi · International Journal of Information Sciences and Techniques · 2012
Synthetic Aperture Radar (SAR) images are inherently affected by multiplicative speckle noise, due to the coherent nature of scattering phenomena.In this paper, a novel algorithm capable of suppressing speckle noise using Particle Swarm Optimization (PSO) technique is presented.The algorithm initially identifies homogenous region from the corrupted image and uses PSO to optimize the Thresholding of curvelet coefficients to recover the original image.Average Power Spectrum Value (APSV) has been used as objective function of PSO.The Proposed algorithm removes Speckle noise effectively and the performance of the algorithm is tested and compared with Mean filter, Median filter, Lee filter, Statistic Lee filter, Kuan filter, frost filter and gamma filter., outperforming conventional filtering methods.