Synthetic aperture radar image de-noising based on Shearlet transform using the context-based model

As the synthetic aperture radar (SAR) has been widely used in agriculture, forestry, hydrology, mining, marine, mapping and other fields, the method to improve the image quality and visual effect of the SAR image has become a hot research issue. The suppression and removal of the speckle of SAR image have become more and more important. This paper analyzes how the noises of the SAR image are generated and the models are appropriate for the characteristics of SAR images. Then based on the advantages of the Shearlet transform, we proposed an SAR image de-noising algorithm which combines the improved Shearlet transform with a cycle spinning de-noising algorithm by using an adaptive threshold method based on the context model. Simulation results show that the proposed algorithm can significantly suppress the speckle noise and improve the peak signal-to-noise ratio (PSNR) of the image; it also holds the characteristics of translational invariance (which can keep the edges of the image detail signal well and inhibit Gibbs phenomenon caused by noise reduction), and it can greatly improve the visual effect.