SAR image edge detection via sparse representation
SAR image edge detection via sparse representation
复制标题
通过稀疏表示进行 SAR 图像边缘检测
DOI：
10.1007/s00500-017-2505-y
复制
发表时间：
2018-04-01
期刊：
影响因子：
4.1
通讯作者：
Zhao, Jie
中科院分区：
文献类型：
--
作者：
Ma, Xiaole;Liu, Shuaiqi;Zhao, Jie
本文提出了一种新的基于稀疏表示去噪算法和一种新的形态学边缘检测器的合成孔径雷达（SAR）图像检测算法。首先对SAR图像进行Shearlet变换得到稀疏表示，然后对Shearlet变换的方向子带系数进行方向形态边缘检测，并通过迭代去噪恢复方向子带系数。最后利用Dempster-Shafer证据理论对各子带边缘进行融合，得到完整的SAR图像边缘。该算法充分利用了Shearlet变换的方向子带，克服了传统的变换检测算法对噪声鲁棒性差、边缘不准确的缺点。实验结果表明，我们提出的算法的有效性和优越性的边缘定位精度，完整性，和虚假边缘点的数量。
In this paper, we propose a new synthetic aperture radar (SAR) image detection algorithm based on the de-noising algorithm via the sparse representation and a new morphology edge detector. Firstly, we apply the Shearlet transform to the SAR image to get the sparse representation of it. Then, morphological edge detector with direction is applied to directional sub-band coefficients of the Shearlet which are recovered by the iterative de-noising process. Finally, the completed SAR image edge is obtained by merging each sub-band edge using Dempster–Shafer evidence theory. By completely using the directional sub-bands of the Shearlet transform, the proposed algorithm overcomes the disadvantages of transform detection algorithms which are very unrobust to noise and can also generate inaccurate edges. The experimental results demonstrate the effectiveness and superiority of our proposed algorithm in terms of the edge positioning accuracy, integrity, and the number of false edge points.