太阳成8722(中国)有限公司-GREEN NO.1

标题: SAR Image Despeckling Employing a Recursive Deep CNN Prior

作者: Shen, HF (Shen, Huanfeng); Zhou, CX (Zhou, Chenxia); Li, J (Li, Jie); Yuan, QQ (Yuan, Qiangqiang)

来源出版物: IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING 卷: 59 期: 1 页: 273-286 DOI: 10.1109/TGRS.2020.2993319 出版年: JAN 2021

摘要: Synthetic aperture radar (SAR) images are inherently affected by speckle noise, for which deep learning-based methods have shown good potential. However, the deep learning-based methods proposed until now directly map low-quality images to high-quality images, and they are unable to characterize the priors for all the kinds of speckle images. The variational method is a classic model optimization approach that establishes the relationship between the clean and noisy images from the perspective of a probability distribution. Therefore, in this article, we propose the recursive deep convolutional neural network (CNN) prior model for SAR image despeckling (SAR-RDCP). First, the data-fitting term and regularization term of the SAR variational model are decoupled into two subproblems, i.e., a data-fitting block and a deep CNN prior block. The gradient descent algorithm is then used to solve the data-fitting block, and a predenoising residual channel attention network based on dilated convolution is used for the deep CNN prior block, which combines an end-to-end iterative optimization training. In the experiments undertaken in this study, the proposed model was compared with several state-of-the-art despeckling methods, obtaining better results in both the quantitative and qualitative evaluations.

入藏号: WOS:000603079000022

语言: English

文献类型: Article

作者关键词: Synthetic aperture radar; Machine learning; Optimization; Speckle; Image restoration; Learning systems; Transforms; Convolutional neural network (CNN); despeckling gain (DG) loss; residual channel attention; synthetic aperture radar (SAR) image despeckling; variation

地址: [Shen, Huanfeng; Zhou, Chenxia] Wuhan Univ, Sch Resource & Environm Sci, Wuhan 430079, Peoples R China.

[Shen, Huanfeng; Yuan, Qiangqiang] Wuhan Univ, Collaborat Innovat Ctr Geospatial Technol, Wuhan 430079, Peoples R China.

[Li, Jie] Wuhan Univ, Sch Geodesy & Geomat, Wuhan 430079, Peoples R China.

[Yuan, Qiangqiang] Sch Geodesy & Geomat, Wuhan 430079, Peoples R China.

通讯作者地址: Li, J (通讯作者)，Wuhan Univ, Sch Geodesy & Geomat, Wuhan 430079, Peoples R China.

电子邮件地址: shenhf@whu.edu.cn; zhoucx31@whu.edu.cn; aaronleecool@whu.edu.cn; yqiang86@gmail.com

影响因子：5.855