Helmet-Wearing Tracking Detection Based on StrongSORT
Object detection based on deep learning is one of the most important and fundamental tasks of computer vision. High-performance detection algorithms have been widely used in many practical fields. For the management of workers wearing helmets in construction scenarios, this paper proposes a framewor...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-02-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/23/3/1682 |
| _version_ | 1797623158796189696 |
|---|---|
| author | Fufang Li Yan Chen Ming Hu Manlin Luo Guobin Wang |
| author_facet | Fufang Li Yan Chen Ming Hu Manlin Luo Guobin Wang |
| author_sort | Fufang Li |
| collection | DOAJ |
| description | Object detection based on deep learning is one of the most important and fundamental tasks of computer vision. High-performance detection algorithms have been widely used in many practical fields. For the management of workers wearing helmets in construction scenarios, this paper proposes a framework model based on the YOLOv5 detection algorithm, combined with multi-object tracking algorithms, to monitor and track whether workers wear safety helmets in real-time video. The improved StrongSORT tracking algorithm of DeepSORT is selected to reduce the loss of the tracked object caused by the occlusion, trajectory blur, and motion scale of the object. The safety helmet dataset is trained with YOLOv5s, and the best result of training is used as the weight model in the StrongSORT tracking algorithm. The experimental results show that the mAP@0.5 of all classes in the YOLOv5s model can reach 95.1% in the validation dataset, mAP@0.5:0.95 is 62.1%, and the precision of wearing helmet is 95.7%. After the box regression loss function was changed from CIOU to Focal-EIOU, the mAP@0.5 increased to 95.4%, mAP@0.5:0.95 increased to 62.9%, and the precision of wearing helmet increased to 96.5%, which were increased by 0.3%, 0.8% and 0.8%, respectively. StrongSORT can update object trajectories in video frames at a speed of 0.05 s per frame. Based on the improved YOLOv5s combined with the StrongSORT tracking algorithm, the helmet-wearing tracking detection can achieve better performance. |
| first_indexed | 2024-03-11T09:24:42Z |
| format | Article |
| id | doaj.art-fae32c53543f4a0498366e4aca825f6d |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-11T09:24:42Z |
| publishDate | 2023-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-fae32c53543f4a0498366e4aca825f6d2023-11-16T18:04:39ZengMDPI AGSensors1424-82202023-02-01233168210.3390/s23031682Helmet-Wearing Tracking Detection Based on StrongSORTFufang Li0Yan Chen1Ming Hu2Manlin Luo3Guobin Wang4School of Computer Science and Cyber Engineering, Guangzhou University, Guangzhou 510006, ChinaSchool of Computer Science and Cyber Engineering, Guangzhou University, Guangzhou 510006, ChinaSchool of Computer Science and Cyber Engineering, Guangzhou University, Guangzhou 510006, ChinaSchool of Computer Science and Cyber Engineering, Guangzhou University, Guangzhou 510006, ChinaSchool of Computer Science and Cyber Engineering, Guangzhou University, Guangzhou 510006, ChinaObject detection based on deep learning is one of the most important and fundamental tasks of computer vision. High-performance detection algorithms have been widely used in many practical fields. For the management of workers wearing helmets in construction scenarios, this paper proposes a framework model based on the YOLOv5 detection algorithm, combined with multi-object tracking algorithms, to monitor and track whether workers wear safety helmets in real-time video. The improved StrongSORT tracking algorithm of DeepSORT is selected to reduce the loss of the tracked object caused by the occlusion, trajectory blur, and motion scale of the object. The safety helmet dataset is trained with YOLOv5s, and the best result of training is used as the weight model in the StrongSORT tracking algorithm. The experimental results show that the mAP@0.5 of all classes in the YOLOv5s model can reach 95.1% in the validation dataset, mAP@0.5:0.95 is 62.1%, and the precision of wearing helmet is 95.7%. After the box regression loss function was changed from CIOU to Focal-EIOU, the mAP@0.5 increased to 95.4%, mAP@0.5:0.95 increased to 62.9%, and the precision of wearing helmet increased to 96.5%, which were increased by 0.3%, 0.8% and 0.8%, respectively. StrongSORT can update object trajectories in video frames at a speed of 0.05 s per frame. Based on the improved YOLOv5s combined with the StrongSORT tracking algorithm, the helmet-wearing tracking detection can achieve better performance.https://www.mdpi.com/1424-8220/23/3/1682YOLOv5focal-EIOUobject detectionStrongSORThelmet wear tracking |
| spellingShingle | Fufang Li Yan Chen Ming Hu Manlin Luo Guobin Wang Helmet-Wearing Tracking Detection Based on StrongSORT Sensors YOLOv5 focal-EIOU object detection StrongSORT helmet wear tracking |
| title | Helmet-Wearing Tracking Detection Based on StrongSORT |
| title_full | Helmet-Wearing Tracking Detection Based on StrongSORT |
| title_fullStr | Helmet-Wearing Tracking Detection Based on StrongSORT |
| title_full_unstemmed | Helmet-Wearing Tracking Detection Based on StrongSORT |
| title_short | Helmet-Wearing Tracking Detection Based on StrongSORT |
| title_sort | helmet wearing tracking detection based on strongsort |
| topic | YOLOv5 focal-EIOU object detection StrongSORT helmet wear tracking |
| url | https://www.mdpi.com/1424-8220/23/3/1682 |
| work_keys_str_mv | AT fufangli helmetwearingtrackingdetectionbasedonstrongsort AT yanchen helmetwearingtrackingdetectionbasedonstrongsort AT minghu helmetwearingtrackingdetectionbasedonstrongsort AT manlinluo helmetwearingtrackingdetectionbasedonstrongsort AT guobinwang helmetwearingtrackingdetectionbasedonstrongsort |