ImageDenoising UsingContour let Transform with Application to Synthetic Aperture Radar

In all methods of image denoising there is a problem always exists that is how to distinguish noise and edge. Now wavelet and contourlet are main tools in image denoising, but threshold is the key in wavelet and contourlet denoising. In order to distinguish noise and edge well, most methods in wavelet denoising are about the improvement of threshold. Aiming to resolve this problem, a new method which makes full use of the model of anisotropic receptive fields and nonsubsampled contourlet transform is proposed and it can distinguish the noise and edge effectively without the need to choose a suitable threshold. Image denoising has become an essential exercise in SAR imaging especially the satellite imaging radars. This paper proposes a SAR image denoising algorithm using contourlet transform. Numerical results show that the proposed algorithm can obtained higher peak signal to noise ratio (PSNR) than wavelet based denoising algorithms using images in the presence of gaussian noise (GN).