SAR image denoising using homomorphic and shearlet transforms
Hossein Rezaei, Azam Karami · 2017
Recently, denoising of Synthetic Aperture Radar (SAR) images has gained particular attention. SAR image is usually affected by speckle noise. In this paper a new method for speckle noise reduction of SAR images using shearlet transform (ST) is introduced. ST could significantly remove the Gaussian noise therefore in the proposed method first, noisy images are converted to a domain which type of noise is Gaussian using homomorphic transform (HT). Second, 2D shear-let is applied to the data. Third, the hard thresholding is used in order to denoise the shearlet coefficients. Finally reconstructed denoised images are obtained by applying the inverse shearlet and homomorphic transforms. The proposed method (ST-HT) is compared with state of art denoising algorithms on SAR images. Obtained results show the superiority of the proposed approach.