A novel computer vision‐based approach for monitoring safety harness use in construction. Issue 4 (22nd November 2022)

• Record Type: Journal Article
• Authors: Xu, Zhijing; Huang, Jiajing; Huang, Kan
• Is Part Of: IET image processing. Volume 17:Issue 4(2023)
• Page Start: 1071
• Page End: 1085
• Publication Date: 2022-11-22
• Subjects: computer vision -- construction industry -- object detection; Image processing -- Periodicals; 621.36705
• DOI: 10.1049/ipr2.12696
• Languages: English
• ISSNs: 1751-9659

Abstract

Abstract: Falling from a height is the most common accident on construction sites. Vision‐based techniques can be used to automatically monitor the construction sites and give early warnings. In this study, a lightweight object detection method, Efficient‐YOLOv5, was proposed for detecting whether workers are wearing safety harnesses when working at height. Furthermore, a matching‐recheck strategy was proposed to improve the mean average precision (mAP). The safety status evaluation model was designed to evaluate the safety status of workers in different construction scenarios. An edge computing‐based security monitoring and alarm system suitable for deployment on construction sites was proposed to assist manual management. Efficient‐YOLOv5 was trained and evaluated on our newly created dataset. Experiments demonstrated that our proposed method outperformed other comparison methods, as the precision and recall rates were 97.7% and 89.3%, respectively. The mAP was 94%. The rate of frames per second (FPS) was 72, which met real‐time application requirements. Thus, the proposed method could easily be applied in the construction industry. Abstract : Falling from a height is the most common accident on construction sites. We proposed a novel computer vision‐based approach to detect workers not wearing harnesses on construction sites. It could assist the construction industry in reducing accidents caused by falling from heights. Experiments demonstrated that our proposed method outperformed other comparison methods, as the precision and recall rates were 97.7% and 89.3%, respectively. The mAP was 94%. The rate of frames per second (FPS) was 72, which met real‐time application requirements.