HPHR-SAR-Net: Hyperpixel High-Resolution SAR Imaging Network Based on Nonlocal Total Variation
High resolution is a key trend in the development of synthetic aperture radar (SAR), which enables the capture of fine details and accurate representation of backscattering properties. However, traditional high-resolution SAR imaging algorithms face several challenges. First, these algorithms tend t...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2023-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10184009/ |
| _version_ | 1797357959016087552 |
|---|---|
| author | Guoru Zhou Zhongqiu Xu Yizhe Fan Zhe Zhang Xiaolan Qiu Bingchen Zhang Kun Fu Yirong Wu |
| author_facet | Guoru Zhou Zhongqiu Xu Yizhe Fan Zhe Zhang Xiaolan Qiu Bingchen Zhang Kun Fu Yirong Wu |
| author_sort | Guoru Zhou |
| collection | DOAJ |
| description | High resolution is a key trend in the development of synthetic aperture radar (SAR), which enables the capture of fine details and accurate representation of backscattering properties. However, traditional high-resolution SAR imaging algorithms face several challenges. First, these algorithms tend to focus on local information, neglecting nonlocal information between different pixel patches. Second, speckle is more pronounced and difficult to filter out in high-resolution SAR images. Third, the process of high-resolution SAR imaging generally involves high time and computational complexity, making real-time imaging difficult to achieve. To address these issues, we propose a hyperpixel high-resolution SAR imaging network (HPHR-SAR-Net) for rapid despeckling in high-resolution modes. Based on the concept of hyperpixel techniques, we initially combine nonconvex and nonlocal total variation as compound regularization. This approach more effectively despeckles and manages the relationship between pixels while reducing bias effects caused by convex constraints. Subsequently, we solve the compound regularization model using the alternating direction method of multipliers (ADMM) algorithm and unfold it into a deep unfolding network (DUN). The network's parameters are adaptively learned in a data-driven manner, and the learned network significantly increases imaging speed. Additionally, the proposed HPHR-SAR-Net is compatible with high-resolution imaging modes, such as spotlight, staring spotlight, and sliding spotlight. In this article, we demonstrate the superiority of HPHR-SAR-Net through experiments in both simulated and real SAR scenarios. The results indicate that HPHR-SAR-Net can rapidly perform high-resolution SAR imaging from raw echo data, producing accurate imaging results. |
| first_indexed | 2024-03-08T14:53:02Z |
| format | Article |
| id | doaj.art-13227a76e24b476482382c1da0c6c9ac |
| institution | Directory Open Access Journal |
| issn | 2151-1535 |
| language | English |
| last_indexed | 2024-03-08T14:53:02Z |
| publishDate | 2023-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj.art-13227a76e24b476482382c1da0c6c9ac2024-01-11T00:00:44ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing2151-15352023-01-01168595860810.1109/JSTARS.2023.329572810184009HPHR-SAR-Net: Hyperpixel High-Resolution SAR Imaging Network Based on Nonlocal Total VariationGuoru Zhou0https://orcid.org/0009-0007-3540-5824Zhongqiu Xu1https://orcid.org/0000-0003-3338-2482Yizhe Fan2https://orcid.org/0009-0005-7551-2275Zhe Zhang3https://orcid.org/0000-0003-3192-3476Xiaolan Qiu4https://orcid.org/0000-0002-8517-3415Bingchen Zhang5Kun Fu6https://orcid.org/0000-0002-0450-6469Yirong Wu7Key Laboratory of Technology in Geo-spatial Information Processing and Application System, Chinese Academy of Sciences, Beijing, ChinaKey Laboratory of Technology in Geo-spatial Information Processing and Application System, Chinese Academy of Sciences, Beijing, ChinaKey Laboratory of Technology in Geo-spatial Information Processing and Application System, Chinese Academy of Sciences, Beijing, ChinaSuzhou Key Laboratory of Microwave Imaging, Processing and Application Technology, Suzhou Aerospace Information Research Institute, Suzhou, ChinaKey Laboratory of Technology in Geo-spatial Information Processing and Application System, Chinese Academy of Sciences, Beijing, ChinaKey Laboratory of Technology in Geo-spatial Information Processing and Application System, Chinese Academy of Sciences, Beijing, ChinaKey Laboratory of Technology in Geo-spatial Information Processing and Application System, Chinese Academy of Sciences, Beijing, ChinaKey Laboratory of Technology in Geo-spatial Information Processing and Application System, Chinese Academy of Sciences, Beijing, ChinaHigh resolution is a key trend in the development of synthetic aperture radar (SAR), which enables the capture of fine details and accurate representation of backscattering properties. However, traditional high-resolution SAR imaging algorithms face several challenges. First, these algorithms tend to focus on local information, neglecting nonlocal information between different pixel patches. Second, speckle is more pronounced and difficult to filter out in high-resolution SAR images. Third, the process of high-resolution SAR imaging generally involves high time and computational complexity, making real-time imaging difficult to achieve. To address these issues, we propose a hyperpixel high-resolution SAR imaging network (HPHR-SAR-Net) for rapid despeckling in high-resolution modes. Based on the concept of hyperpixel techniques, we initially combine nonconvex and nonlocal total variation as compound regularization. This approach more effectively despeckles and manages the relationship between pixels while reducing bias effects caused by convex constraints. Subsequently, we solve the compound regularization model using the alternating direction method of multipliers (ADMM) algorithm and unfold it into a deep unfolding network (DUN). The network's parameters are adaptively learned in a data-driven manner, and the learned network significantly increases imaging speed. Additionally, the proposed HPHR-SAR-Net is compatible with high-resolution imaging modes, such as spotlight, staring spotlight, and sliding spotlight. In this article, we demonstrate the superiority of HPHR-SAR-Net through experiments in both simulated and real SAR scenarios. The results indicate that HPHR-SAR-Net can rapidly perform high-resolution SAR imaging from raw echo data, producing accurate imaging results.https://ieeexplore.ieee.org/document/10184009/Alternating direction method of multipliers (ADMM)deep unfolding network (DUN)high resolutionhyperpixelsparse microwave imagingsynthetic aperture radar (SAR) |
| spellingShingle | Guoru Zhou Zhongqiu Xu Yizhe Fan Zhe Zhang Xiaolan Qiu Bingchen Zhang Kun Fu Yirong Wu HPHR-SAR-Net: Hyperpixel High-Resolution SAR Imaging Network Based on Nonlocal Total Variation IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Alternating direction method of multipliers (ADMM) deep unfolding network (DUN) high resolution hyperpixel sparse microwave imaging synthetic aperture radar (SAR) |
| title | HPHR-SAR-Net: Hyperpixel High-Resolution SAR Imaging Network Based on Nonlocal Total Variation |
| title_full | HPHR-SAR-Net: Hyperpixel High-Resolution SAR Imaging Network Based on Nonlocal Total Variation |
| title_fullStr | HPHR-SAR-Net: Hyperpixel High-Resolution SAR Imaging Network Based on Nonlocal Total Variation |
| title_full_unstemmed | HPHR-SAR-Net: Hyperpixel High-Resolution SAR Imaging Network Based on Nonlocal Total Variation |
| title_short | HPHR-SAR-Net: Hyperpixel High-Resolution SAR Imaging Network Based on Nonlocal Total Variation |
| title_sort | hphr sar net hyperpixel high resolution sar imaging network based on nonlocal total variation |
| topic | Alternating direction method of multipliers (ADMM) deep unfolding network (DUN) high resolution hyperpixel sparse microwave imaging synthetic aperture radar (SAR) |
| url | https://ieeexplore.ieee.org/document/10184009/ |
| work_keys_str_mv | AT guoruzhou hphrsarnethyperpixelhighresolutionsarimagingnetworkbasedonnonlocaltotalvariation AT zhongqiuxu hphrsarnethyperpixelhighresolutionsarimagingnetworkbasedonnonlocaltotalvariation AT yizhefan hphrsarnethyperpixelhighresolutionsarimagingnetworkbasedonnonlocaltotalvariation AT zhezhang hphrsarnethyperpixelhighresolutionsarimagingnetworkbasedonnonlocaltotalvariation AT xiaolanqiu hphrsarnethyperpixelhighresolutionsarimagingnetworkbasedonnonlocaltotalvariation AT bingchenzhang hphrsarnethyperpixelhighresolutionsarimagingnetworkbasedonnonlocaltotalvariation AT kunfu hphrsarnethyperpixelhighresolutionsarimagingnetworkbasedonnonlocaltotalvariation AT yirongwu hphrsarnethyperpixelhighresolutionsarimagingnetworkbasedonnonlocaltotalvariation |