Synthetic-Aperture Radar Image Despeckling based on Improved Non-Local Means and Non-Subsampled Shearlet Transform

Zengguo Sun,Rui Shi,Wei Wei

doi:10.5755/j01.itc.49.3.23998

Abstract

When Synthetic-Aperture (SAR) image is transformed into wavelet domain and other transform domains, most of the coefficients of the image are small or zero. This shows that SAR image is sparse. However, speckle can be seen in SAR images. The non-local means is a despeckling algorithm, but it cannot overcome the speckle in homogeneous regions and it blurs edge details of the image. In order to solve these problems, an improved non-local means is suggested. At the same time, in order to better suppress the speckle effectively in edge regions, the non-subsampled Shearlet transform (NSST) is applied. By combining NSST with the improved non-local means, a new type of despeckling algorithm is proposed. Results show that the proposed algorithm leads to a satisfying performance for SAR images.

Highlights

The GF-3 is the first C-band and multi-polarized synthetic-aperture radar (SAR) satellite of 1-metre resolution [36]
When SAR image is transformed into wavelet domain and other transform domains, most of the transformed coefficients of the image are small or zero, i.e., the SAR image is sparse
To better suppress the speckle in edge regions in SAR images, Shearlet is used for the construction of the new despeckling algorithm

Summary

Introduction

The GF-3 is the first C-band and multi-polarized synthetic-aperture radar (SAR) satellite of 1-metre resolution [36]. For the GF-3 images, positions and geometric structures such as point targets, textures and edges are so clear, which indicates that GF-3 images have clear sparsity The use of this sparsity is helpful for construction of efficient despeckling methods. It should be noted that the improved NLM cannot smooth speckle sufficiently in edge regions This drawback can be eliminated by Shearlet. To better suppress the speckle in edge regions in SAR images, Shearlet is used for the construction of the new despeckling algorithm. Our proposed method combines the improved NLM with NSST, so it can sufficiently reflect the sparsity of SAR images. The proposed method reduces speckle in homogeneous areas, effectively preserves edge information, and better smoothes speckle in edge areas, leading to a satisfying performance for SAR images

Brief Introduction of SAR Images

Proposed Despeckling Method

Conclusion