A Joint Moving Target Detection Method in Video SAR Via Low-Rank Sparse Decomposition and Transformer
Video synthetic aperture radar (SAR) has exhibited considerable potential for detecting and tracking ground moving targets. Numerous classical shadow-based detection methods have been applied in video SAR. In addition, shadow-assisted detection methods based on convolutional neural networks (CNNs) h...
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
IEEE
2025-01-01
|
| Series: | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| Subjects: | |
| Online Access: | https://ieeexplore.ieee.org/document/10759747/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1846129710508015616 |
|---|---|
| author | Hui Fang Guisheng Liao Yongjun Liu Cao Zeng Xiongpeng He Mingming Xu |
| author_facet | Hui Fang Guisheng Liao Yongjun Liu Cao Zeng Xiongpeng He Mingming Xu |
| author_sort | Hui Fang |
| collection | DOAJ |
| description | Video synthetic aperture radar (SAR) has exhibited considerable potential for detecting and tracking ground moving targets. Numerous classical shadow-based detection methods have been applied in video SAR. In addition, shadow-assisted detection methods based on convolutional neural networks (CNNs) have been developed. In this article, we propose a joint detection method for moving targets in video SAR, which can combine the information of the video SAR image and the corresponding sparse image to suppress background interference sufficiently and improve detection accuracy. Specifically, the low-rank sparse decomposition technology is first applied for video SAR images to generate their corresponding sparse images in which the background is eliminated and shadows of moving targets are enhanced. Then, we improve faster RCNN and build a two-stream extraction feature network based on the Transformer structure that allows the video SAR image and the sparse image as input simultaneously as well as extracts and fuses the features from two types of the images, which can acquire more discriminative target features, thereby improving the final the detection performance. Furthermore, the improved faster RCNN only modifies the original feature extraction network. Thus, it can adopt the same training and test manner as faster RCNN, greatly facilitating its utilization. Finally, experiment results on Sandia National Laboratories data demonstrate that the proposed detection method outperforms other state-of-the-art methods. And our method reduces the false alarms by 1.02%, the missed detections by 43.24%, and increases the mean average precision by 2.98%. |
| format | Article |
| id | doaj-art-c3f0cd13ac784a8a88e007467f52ada0 |
| institution | Kabale University |
| issn | 1939-1404 2151-1535 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | IEEE |
| record_format | Article |
| series | IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing |
| spelling | doaj-art-c3f0cd13ac784a8a88e007467f52ada02024-12-10T00:00:33ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-01181007101910.1109/JSTARS.2024.350363910759747A Joint Moving Target Detection Method in Video SAR Via Low-Rank Sparse Decomposition and TransformerHui Fang0https://orcid.org/0000-0002-1902-8156Guisheng Liao1https://orcid.org/0000-0002-5919-0713Yongjun Liu2https://orcid.org/0000-0003-1628-8865Cao Zeng3https://orcid.org/0000-0001-5842-3629Xiongpeng He4https://orcid.org/0000-0001-8873-8998Mingming Xu5https://orcid.org/0000-0002-7159-6994National Key Laboratory of Radar Signal Processing, Xidian University, Xi'an, ChinaNational Key Laboratory of Radar Signal Processing, Xidian University, Xi'an, ChinaNational Key Laboratory of Radar Signal Processing, Xidian University, Xi'an, ChinaNational Key Laboratory of Radar Signal Processing, Xidian University, Xi'an, ChinaNational Key Laboratory of Radar Signal Processing, Xidian University, Xi'an, ChinaInstitute of Remote Sensing Satellite, China Academy of Space Technology, Beijing, ChinaVideo synthetic aperture radar (SAR) has exhibited considerable potential for detecting and tracking ground moving targets. Numerous classical shadow-based detection methods have been applied in video SAR. In addition, shadow-assisted detection methods based on convolutional neural networks (CNNs) have been developed. In this article, we propose a joint detection method for moving targets in video SAR, which can combine the information of the video SAR image and the corresponding sparse image to suppress background interference sufficiently and improve detection accuracy. Specifically, the low-rank sparse decomposition technology is first applied for video SAR images to generate their corresponding sparse images in which the background is eliminated and shadows of moving targets are enhanced. Then, we improve faster RCNN and build a two-stream extraction feature network based on the Transformer structure that allows the video SAR image and the sparse image as input simultaneously as well as extracts and fuses the features from two types of the images, which can acquire more discriminative target features, thereby improving the final the detection performance. Furthermore, the improved faster RCNN only modifies the original feature extraction network. Thus, it can adopt the same training and test manner as faster RCNN, greatly facilitating its utilization. Finally, experiment results on Sandia National Laboratories data demonstrate that the proposed detection method outperforms other state-of-the-art methods. And our method reduces the false alarms by 1.02%, the missed detections by 43.24%, and increases the mean average precision by 2.98%.https://ieeexplore.ieee.org/document/10759747/Convolutional neural network (CNN)low-rank sparse decomposition (LSD)shadowtarget detectionvideo synthetic aperture radar (video SAR) |
| spellingShingle | Hui Fang Guisheng Liao Yongjun Liu Cao Zeng Xiongpeng He Mingming Xu A Joint Moving Target Detection Method in Video SAR Via Low-Rank Sparse Decomposition and Transformer IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Convolutional neural network (CNN) low-rank sparse decomposition (LSD) shadow target detection video synthetic aperture radar (video SAR) |
| title | A Joint Moving Target Detection Method in Video SAR Via Low-Rank Sparse Decomposition and Transformer |
| title_full | A Joint Moving Target Detection Method in Video SAR Via Low-Rank Sparse Decomposition and Transformer |
| title_fullStr | A Joint Moving Target Detection Method in Video SAR Via Low-Rank Sparse Decomposition and Transformer |
| title_full_unstemmed | A Joint Moving Target Detection Method in Video SAR Via Low-Rank Sparse Decomposition and Transformer |
| title_short | A Joint Moving Target Detection Method in Video SAR Via Low-Rank Sparse Decomposition and Transformer |
| title_sort | joint moving target detection method in video sar via low rank sparse decomposition and transformer |
| topic | Convolutional neural network (CNN) low-rank sparse decomposition (LSD) shadow target detection video synthetic aperture radar (video SAR) |
| url | https://ieeexplore.ieee.org/document/10759747/ |
| work_keys_str_mv | AT huifang ajointmovingtargetdetectionmethodinvideosarvialowranksparsedecompositionandtransformer AT guishengliao ajointmovingtargetdetectionmethodinvideosarvialowranksparsedecompositionandtransformer AT yongjunliu ajointmovingtargetdetectionmethodinvideosarvialowranksparsedecompositionandtransformer AT caozeng ajointmovingtargetdetectionmethodinvideosarvialowranksparsedecompositionandtransformer AT xiongpenghe ajointmovingtargetdetectionmethodinvideosarvialowranksparsedecompositionandtransformer AT mingmingxu ajointmovingtargetdetectionmethodinvideosarvialowranksparsedecompositionandtransformer AT huifang jointmovingtargetdetectionmethodinvideosarvialowranksparsedecompositionandtransformer AT guishengliao jointmovingtargetdetectionmethodinvideosarvialowranksparsedecompositionandtransformer AT yongjunliu jointmovingtargetdetectionmethodinvideosarvialowranksparsedecompositionandtransformer AT caozeng jointmovingtargetdetectionmethodinvideosarvialowranksparsedecompositionandtransformer AT xiongpenghe jointmovingtargetdetectionmethodinvideosarvialowranksparsedecompositionandtransformer AT mingmingxu jointmovingtargetdetectionmethodinvideosarvialowranksparsedecompositionandtransformer |