A method for removing speckles of SAR image based on markov random field
Jun Cheng · Journal of Sichuan University · 2008
In synthetic aperture radar(SAR) imaging,the speckles affect seriously on the quality of SAR images.Because Markov random field(MRF) can be used to describe the spatial continuity,the structures of images can be modeled if a proper neighborhood system is selected.In order to remove the speckles,therefore,Gibbs-MRF and simulated annealing are proposed to remove speckles of SAR images.The intensities of the images can be considered as the radar cross-section(RCS) areas,a simulated annealing algorithm,combined with Bayesian criteria for estimation of the radar cross-section areas,is developed to remove speckles.Experiment results show that the method proposed here is effective.