SDES-YOLO: A high-precision and lightweight model for fall detection in complex environments
Abstract Falling is an emergency situation that can result in serious injury or even death, especially in the absence of immediate assistance. Therefore, developing a model that can accurately and promptly detect falls is crucial for enhancing quality of life and safety. In the field of object detec...
Saved in:
| Main Authors: | , , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-01-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-025-86593-9 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832594699189223424 |
|---|---|
| author | Xiangqian Huang Xiaoming Li Limengzi Yuan Zhao Jiang Hongwei Jin Wanghao Wu Ru Cai Meilian Zheng Hongpeng Bai |
| author_facet | Xiangqian Huang Xiaoming Li Limengzi Yuan Zhao Jiang Hongwei Jin Wanghao Wu Ru Cai Meilian Zheng Hongpeng Bai |
| author_sort | Xiangqian Huang |
| collection | DOAJ |
| description | Abstract Falling is an emergency situation that can result in serious injury or even death, especially in the absence of immediate assistance. Therefore, developing a model that can accurately and promptly detect falls is crucial for enhancing quality of life and safety. In the field of object detection, while YOLOv8 has recently made notable strides in detection accuracy and speed, it still faces challenges in detecting falls due to variations in lighting, occlusions, and complex human postures. To address these issues, this study proposes the SDES-YOLO model, an improvement based on YOLOv8. By incorporating a multi-scale feature extraction pyramid (SDFP), occlusion-aware attention mechanism (SEAM), an edge and spatial information fusion module (ES3), and a WIoU-Shape loss function, the SDES-YOLO model significantly enhances fall detection performance in complex scenarios. With only 2.9M parameters and 7.2 GFLOPs of computation, SDES-YOLO achieves an mAP@0.5 of 85.1%, representing a 3.41% improvement over YOLOv8n, while reducing parameter count and computation by 1.33% and 11.11%, respectively. These results indicate that SDES-YOLO successfully combines efficiency and precision in fall detection. Through these innovations, SDES-YOLO not only improves detection accuracy but also optimizes computational efficiency, making it effective even in resource-constrained environments. |
| format | Article |
| id | doaj-art-7b2dd54454d8403f9c956dc14a44d833 |
| institution | Kabale University |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-01-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-7b2dd54454d8403f9c956dc14a44d8332025-01-19T12:24:13ZengNature PortfolioScientific Reports2045-23222025-01-0115111810.1038/s41598-025-86593-9SDES-YOLO: A high-precision and lightweight model for fall detection in complex environmentsXiangqian Huang0Xiaoming Li1Limengzi Yuan2Zhao Jiang3Hongwei Jin4Wanghao Wu5Ru Cai6Meilian Zheng7Hongpeng Bai8International Business School, Zhejiang Yuexiu UniversityInternational Business School, Zhejiang Yuexiu UniversityCollege of Information Science and Technology, Shihezi UniversityInternational Business School, Zhejiang Yuexiu UniversityInternational Business School, Zhejiang Yuexiu UniversityInternational Business School, Zhejiang Yuexiu UniversityInternational Business School, Zhejiang Yuexiu UniversitySchool of Management, Zhejiang University of TechnologyCollege of Intelligence and Computing, Tianjin UniversityAbstract Falling is an emergency situation that can result in serious injury or even death, especially in the absence of immediate assistance. Therefore, developing a model that can accurately and promptly detect falls is crucial for enhancing quality of life and safety. In the field of object detection, while YOLOv8 has recently made notable strides in detection accuracy and speed, it still faces challenges in detecting falls due to variations in lighting, occlusions, and complex human postures. To address these issues, this study proposes the SDES-YOLO model, an improvement based on YOLOv8. By incorporating a multi-scale feature extraction pyramid (SDFP), occlusion-aware attention mechanism (SEAM), an edge and spatial information fusion module (ES3), and a WIoU-Shape loss function, the SDES-YOLO model significantly enhances fall detection performance in complex scenarios. With only 2.9M parameters and 7.2 GFLOPs of computation, SDES-YOLO achieves an mAP@0.5 of 85.1%, representing a 3.41% improvement over YOLOv8n, while reducing parameter count and computation by 1.33% and 11.11%, respectively. These results indicate that SDES-YOLO successfully combines efficiency and precision in fall detection. Through these innovations, SDES-YOLO not only improves detection accuracy but also optimizes computational efficiency, making it effective even in resource-constrained environments.https://doi.org/10.1038/s41598-025-86593-9Fall detectionMulti-scale feature extraction pyramidOcclusion-aware attention mechanismEdge and spatial information fusion moduleWIoU-Shape loss functionYOLO |
| spellingShingle | Xiangqian Huang Xiaoming Li Limengzi Yuan Zhao Jiang Hongwei Jin Wanghao Wu Ru Cai Meilian Zheng Hongpeng Bai SDES-YOLO: A high-precision and lightweight model for fall detection in complex environments Scientific Reports Fall detection Multi-scale feature extraction pyramid Occlusion-aware attention mechanism Edge and spatial information fusion module WIoU-Shape loss function YOLO |
| title | SDES-YOLO: A high-precision and lightweight model for fall detection in complex environments |
| title_full | SDES-YOLO: A high-precision and lightweight model for fall detection in complex environments |
| title_fullStr | SDES-YOLO: A high-precision and lightweight model for fall detection in complex environments |
| title_full_unstemmed | SDES-YOLO: A high-precision and lightweight model for fall detection in complex environments |
| title_short | SDES-YOLO: A high-precision and lightweight model for fall detection in complex environments |
| title_sort | sdes yolo a high precision and lightweight model for fall detection in complex environments |
| topic | Fall detection Multi-scale feature extraction pyramid Occlusion-aware attention mechanism Edge and spatial information fusion module WIoU-Shape loss function YOLO |
| url | https://doi.org/10.1038/s41598-025-86593-9 |
| work_keys_str_mv | AT xiangqianhuang sdesyoloahighprecisionandlightweightmodelforfalldetectionincomplexenvironments AT xiaomingli sdesyoloahighprecisionandlightweightmodelforfalldetectionincomplexenvironments AT limengziyuan sdesyoloahighprecisionandlightweightmodelforfalldetectionincomplexenvironments AT zhaojiang sdesyoloahighprecisionandlightweightmodelforfalldetectionincomplexenvironments AT hongweijin sdesyoloahighprecisionandlightweightmodelforfalldetectionincomplexenvironments AT wanghaowu sdesyoloahighprecisionandlightweightmodelforfalldetectionincomplexenvironments AT rucai sdesyoloahighprecisionandlightweightmodelforfalldetectionincomplexenvironments AT meilianzheng sdesyoloahighprecisionandlightweightmodelforfalldetectionincomplexenvironments AT hongpengbai sdesyoloahighprecisionandlightweightmodelforfalldetectionincomplexenvironments |