Self-Supervised Despeckling Algorithm with an Enhanced U-Net for Synthetic Aperture Radar Images
Self-supervised method has proven to be a suitable approach for despeckling on synthetic aperture radar (SAR) images. However, most self-supervised despeckling methods are trained by noisy-noisy image pairs, which are constructed by using natural images with simulated speckle noise, time-series real...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-10-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/13/21/4383 |
| _version_ | 1827677969707958272 |
|---|---|
| author | Gang Zhang Zhi Li Xuewei Li Sitong Liu |
| author_facet | Gang Zhang Zhi Li Xuewei Li Sitong Liu |
| author_sort | Gang Zhang |
| collection | DOAJ |
| description | Self-supervised method has proven to be a suitable approach for despeckling on synthetic aperture radar (SAR) images. However, most self-supervised despeckling methods are trained by noisy-noisy image pairs, which are constructed by using natural images with simulated speckle noise, time-series real-world SAR images or generative adversarial network, limiting the practicability of these methods in real-world SAR images. Therefore, in this paper, a novel self-supervised despeckling algorithm with an enhanced U-Net is proposed for real-world SAR images. Firstly, unlike previous self-supervised despeckling works, the noisy-noisy image pairs are generated from real-word SAR images through a novel generation training pairs module, which makes it possible to train deep convolutional neural networks using real-world SAR images. Secondly, an enhanced U-Net is designed to improve the feature extraction and fusion capabilities of the network. Thirdly, a self-supervised training loss function with a regularization loss is proposed to address the difference of target pixel values between neighbors on the original SAR images. Finally, visual and quantitative experiments on simulated and real-world SAR images show that the proposed algorithm notably removes speckle noise with better preserving features, which exceed several state-of-the-art despeckling methods. |
| first_indexed | 2024-03-10T05:54:08Z |
| format | Article |
| id | doaj.art-af7902b8d2054e7d8b7a51424ef8c487 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T05:54:08Z |
| publishDate | 2021-10-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-af7902b8d2054e7d8b7a51424ef8c4872023-11-22T21:32:43ZengMDPI AGRemote Sensing2072-42922021-10-011321438310.3390/rs13214383Self-Supervised Despeckling Algorithm with an Enhanced U-Net for Synthetic Aperture Radar ImagesGang Zhang0Zhi Li1Xuewei Li2Sitong Liu3Department of Aerospace Science and Technology, Space Engineering University, Bayi Road, Huairou District, Beijing 101416, ChinaDepartment of Aerospace Science and Technology, Space Engineering University, Bayi Road, Huairou District, Beijing 101416, ChinaInstitute of Software, Chinese Academy of Sciences, No. 4, South Fourth Street, Zhongguancun, Haidian District, Beijing 100190, ChinaDepartment of Aerospace Science and Technology, Space Engineering University, Bayi Road, Huairou District, Beijing 101416, ChinaSelf-supervised method has proven to be a suitable approach for despeckling on synthetic aperture radar (SAR) images. However, most self-supervised despeckling methods are trained by noisy-noisy image pairs, which are constructed by using natural images with simulated speckle noise, time-series real-world SAR images or generative adversarial network, limiting the practicability of these methods in real-world SAR images. Therefore, in this paper, a novel self-supervised despeckling algorithm with an enhanced U-Net is proposed for real-world SAR images. Firstly, unlike previous self-supervised despeckling works, the noisy-noisy image pairs are generated from real-word SAR images through a novel generation training pairs module, which makes it possible to train deep convolutional neural networks using real-world SAR images. Secondly, an enhanced U-Net is designed to improve the feature extraction and fusion capabilities of the network. Thirdly, a self-supervised training loss function with a regularization loss is proposed to address the difference of target pixel values between neighbors on the original SAR images. Finally, visual and quantitative experiments on simulated and real-world SAR images show that the proposed algorithm notably removes speckle noise with better preserving features, which exceed several state-of-the-art despeckling methods.https://www.mdpi.com/2072-4292/13/21/4383self-supervisedsynthetic aperture radar (SAR)despecklingenhanced U-Net |
| spellingShingle | Gang Zhang Zhi Li Xuewei Li Sitong Liu Self-Supervised Despeckling Algorithm with an Enhanced U-Net for Synthetic Aperture Radar Images Remote Sensing self-supervised synthetic aperture radar (SAR) despeckling enhanced U-Net |
| title | Self-Supervised Despeckling Algorithm with an Enhanced U-Net for Synthetic Aperture Radar Images |
| title_full | Self-Supervised Despeckling Algorithm with an Enhanced U-Net for Synthetic Aperture Radar Images |
| title_fullStr | Self-Supervised Despeckling Algorithm with an Enhanced U-Net for Synthetic Aperture Radar Images |
| title_full_unstemmed | Self-Supervised Despeckling Algorithm with an Enhanced U-Net for Synthetic Aperture Radar Images |
| title_short | Self-Supervised Despeckling Algorithm with an Enhanced U-Net for Synthetic Aperture Radar Images |
| title_sort | self supervised despeckling algorithm with an enhanced u net for synthetic aperture radar images |
| topic | self-supervised synthetic aperture radar (SAR) despeckling enhanced U-Net |
| url | https://www.mdpi.com/2072-4292/13/21/4383 |
| work_keys_str_mv | AT gangzhang selfsuperviseddespecklingalgorithmwithanenhancedunetforsyntheticapertureradarimages AT zhili selfsuperviseddespecklingalgorithmwithanenhancedunetforsyntheticapertureradarimages AT xueweili selfsuperviseddespecklingalgorithmwithanenhancedunetforsyntheticapertureradarimages AT sitongliu selfsuperviseddespecklingalgorithmwithanenhancedunetforsyntheticapertureradarimages |