ABSTRACT
The wavelet de-noising related algorithm was proposed by Xu [2] (Spatially Selective Noise Filter, SSNF) and is a denoising algorithm based on edge detection of a representative. Contourlet transform has a similar characteristics, so the algorithm can also be extended to the domain of contourlet transform. After the outline of image wavelet transform coefficients after quite sparse, the large amplitude coefficient is relatively small, often gathered in the vicinity of the contour edges, and in between there is a certain correlation between the scale and continuity, the magnitude of most of the coefficient is close to 0[3].
