A SAR Image Despeckling Using Self-adaptive Threshold FunctionBased on NSCT
Gong Zhang · Journal of Jiamusi University · 2009
A new algorithm which was an energy self-adaptive threshold function for the SAR image denoising was presented in this paper.This algorithm avoided the shortcomings of the hard-threshold function and the soft-threshold function.It combined the shift-invariance with direction selectivity of the nonsubsampled contourlet transform,and also accommodated the threshold function parameter itself by consideration of the image details which have different energy in each subbands and directions.The results of the expeirments on the SAR image denoisnig show that: the proposed algorithm can reduce the speckle noise more effectively while preserving the edges of the SAR image.