ASCDet: cross-space UAV object detection method guided by adaptive sparse convolution
UAV object detection, a critical aspect of remote sensing applications, faces challenges due to high object sparsity and complex backgrounds, leading to excessive computational demands. To address these issues, we propose the Cross-Space UAV Object Detection Method Guided by Adaptive Sparse Convolut...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2025-08-01
|
| Series: | International Journal of Digital Earth |
| Subjects: | |
| Online Access: | https://www.tandfonline.com/doi/10.1080/17538947.2025.2528648 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849224271355707392 |
|---|---|
| author | Gui Cheng Xubin Feng Yan Tian Meilin Xie Chaoya Dang Qing Ding Zhenfeng Shao |
| author_facet | Gui Cheng Xubin Feng Yan Tian Meilin Xie Chaoya Dang Qing Ding Zhenfeng Shao |
| author_sort | Gui Cheng |
| collection | DOAJ |
| description | UAV object detection, a critical aspect of remote sensing applications, faces challenges due to high object sparsity and complex backgrounds, leading to excessive computational demands. To address these issues, we propose the Cross-Space UAV Object Detection Method Guided by Adaptive Sparse Convolution (ASCDet), a more efficient solution for UAV remote sensing object detection. ASCDet introduces a plug-and-play detection head compatible with various detection frameworks, significantly reducing computational costs. The method utilizes an adaptive pixel-level mask unit based on a task-alignment strategy to accurately localize object regions. These masks guide cross-space object detection through sparse convolutions, while a global context enhancement strategy within the sparse convolution module enriches the contextual information, maintaining detection accuracy. Extensive experiments on the VisDrone and UAVDT datasets, comparing ASCDet with benchmark methods such as Faster R-CNN, RetinaNet, FSAF, GFL V1, and TOOD, show that ASCDet improves AP[Formula: see text] by up to 5.8%, increases FPS by up to 34.3%, and reduces GFLOPs by approximately 80%. Additionally, ASCDet enhances detection performance for anchor-free methods, demonstrating superior accuracy and computational efficiency. These results highlight ASCDet's effectiveness in improving detection accuracy, computational efficiency, and sparse detection performance in UAV remote sensing tasks. |
| format | Article |
| id | doaj-art-216602451b054f81b62d8a8e9fe59d63 |
| institution | Kabale University |
| issn | 1753-8947 1753-8955 |
| language | English |
| publishDate | 2025-08-01 |
| publisher | Taylor & Francis Group |
| record_format | Article |
| series | International Journal of Digital Earth |
| spelling | doaj-art-216602451b054f81b62d8a8e9fe59d632025-08-25T11:32:05ZengTaylor & Francis GroupInternational Journal of Digital Earth1753-89471753-89552025-08-0118110.1080/17538947.2025.2528648ASCDet: cross-space UAV object detection method guided by adaptive sparse convolutionGui Cheng0Xubin Feng1Yan Tian2Meilin Xie3Chaoya Dang4Qing Ding5Zhenfeng Shao6Key Laboratory of Space Precision Measurement Technology, Xi'an Institute of Optics and Precision Mechanics, Chinese Academy of Sciences, Xi'an, People's Republic of ChinaKey Laboratory of Space Precision Measurement Technology, Xi'an Institute of Optics and Precision Mechanics, Chinese Academy of Sciences, Xi'an, People's Republic of ChinaKey Laboratory of Space Precision Measurement Technology, Xi'an Institute of Optics and Precision Mechanics, Chinese Academy of Sciences, Xi'an, People's Republic of ChinaKey Laboratory of Space Precision Measurement Technology, Xi'an Institute of Optics and Precision Mechanics, Chinese Academy of Sciences, Xi'an, People's Republic of ChinaThe College of Agriculture, Nanjing Agricultural University, Nanjing, People's Republic of ChinaThe College of Geo-Exploration Science and Technology, Jilin University, Changchun, People's Republic of ChinaThe State Key Laboratory of Information Engineering in Surveying, Mapping and Remote Sensing, Wuhan University, Wuhan, People's Republic of ChinaUAV object detection, a critical aspect of remote sensing applications, faces challenges due to high object sparsity and complex backgrounds, leading to excessive computational demands. To address these issues, we propose the Cross-Space UAV Object Detection Method Guided by Adaptive Sparse Convolution (ASCDet), a more efficient solution for UAV remote sensing object detection. ASCDet introduces a plug-and-play detection head compatible with various detection frameworks, significantly reducing computational costs. The method utilizes an adaptive pixel-level mask unit based on a task-alignment strategy to accurately localize object regions. These masks guide cross-space object detection through sparse convolutions, while a global context enhancement strategy within the sparse convolution module enriches the contextual information, maintaining detection accuracy. Extensive experiments on the VisDrone and UAVDT datasets, comparing ASCDet with benchmark methods such as Faster R-CNN, RetinaNet, FSAF, GFL V1, and TOOD, show that ASCDet improves AP[Formula: see text] by up to 5.8%, increases FPS by up to 34.3%, and reduces GFLOPs by approximately 80%. Additionally, ASCDet enhances detection performance for anchor-free methods, demonstrating superior accuracy and computational efficiency. These results highlight ASCDet's effectiveness in improving detection accuracy, computational efficiency, and sparse detection performance in UAV remote sensing tasks.https://www.tandfonline.com/doi/10.1080/17538947.2025.2528648UAVsparse object detectionadaptive pixel-level maskglobal context enhancement |
| spellingShingle | Gui Cheng Xubin Feng Yan Tian Meilin Xie Chaoya Dang Qing Ding Zhenfeng Shao ASCDet: cross-space UAV object detection method guided by adaptive sparse convolution International Journal of Digital Earth UAV sparse object detection adaptive pixel-level mask global context enhancement |
| title | ASCDet: cross-space UAV object detection method guided by adaptive sparse convolution |
| title_full | ASCDet: cross-space UAV object detection method guided by adaptive sparse convolution |
| title_fullStr | ASCDet: cross-space UAV object detection method guided by adaptive sparse convolution |
| title_full_unstemmed | ASCDet: cross-space UAV object detection method guided by adaptive sparse convolution |
| title_short | ASCDet: cross-space UAV object detection method guided by adaptive sparse convolution |
| title_sort | ascdet cross space uav object detection method guided by adaptive sparse convolution |
| topic | UAV sparse object detection adaptive pixel-level mask global context enhancement |
| url | https://www.tandfonline.com/doi/10.1080/17538947.2025.2528648 |
| work_keys_str_mv | AT guicheng ascdetcrossspaceuavobjectdetectionmethodguidedbyadaptivesparseconvolution AT xubinfeng ascdetcrossspaceuavobjectdetectionmethodguidedbyadaptivesparseconvolution AT yantian ascdetcrossspaceuavobjectdetectionmethodguidedbyadaptivesparseconvolution AT meilinxie ascdetcrossspaceuavobjectdetectionmethodguidedbyadaptivesparseconvolution AT chaoyadang ascdetcrossspaceuavobjectdetectionmethodguidedbyadaptivesparseconvolution AT qingding ascdetcrossspaceuavobjectdetectionmethodguidedbyadaptivesparseconvolution AT zhenfengshao ascdetcrossspaceuavobjectdetectionmethodguidedbyadaptivesparseconvolution |