GF-3 SAR IMAGE DESPECKLING BASED ON THE IMPROVED NON-LOCAL MEANS USING NON-SUBSAMPLED SHEARLET TRANSFORM
GF-3 synthetic aperture radar(SAR)images are rich in information and have obvious sparse features.However,the speckle appears in the GF-3 SAR images due to the coherent imaging system and it hinders the interpretation of images seriously.Recently,Shearlet is applied to the image processing with its best sparse representation.A new Shearlet-transform-based method is proposed in this paper based on the improved non-local means.Firstly,the logarithmic operation and the non-subsampled Shearlet transformation are applied to the GF-3 SAR image.Secondly,in order to solve the problems that the image details are smoothed overly and the weight distribution is affected by the speckle,a new non-local means is used for the transformed high frequency coefficient.Thirdly,the Shearlet reconstruction is carried out.Finally,the final filtered image is obtained by an exponential operation.Experimental results demonstrate that,compared with other despeckling methods,the proposed method can suppress the speckle effectively in homogeneous regions and has better capability of edge preserving.
GF-3 SAR non-subsampled Shearlet transform image despeckling improved Non-Local Means
Rui Shi Zengguo Sun
School of Computer Science,Shaanxi Normal University,Xian China
国际会议
北京
英文
1547-1551
2018-05-07（万方平台首次上网日期，不代表论文的发表时间）