Research on Intrusion Identification of Hazardous Construction Areas Based on Machine Vision. Issue 1 (1st February 2023)

• Record Type: Journal Article
• Authors: Hu, Rongyi; Li, Feng; Wang, Tingting; Fan, Haifeng; Dong, Long
• Is Part Of: Journal of physics. Volume 2435 Issue 1(2023)
• Publication Date: 2023-02-01
• Subjects: Physics -- Congresses; 530.5
• DOI: 10.1088/1742-6596/2435/1/012015
• Languages: English
• ISSNs: 1742-6588

Abstract

Abstract: The complex environment of power construction sites is prone to frequent accidents and many dangerous areas. Using advanced artificial intelligence technology to monitor these dangerous areas in real-time is the primary measure to improve construction safety. Given the problems of poor real-time monitoring and the low accuracy of traditional methods, this paper proposes a real-time monitoring method for dangerous areas based on YOLOv5 deep learning, and through the trained YOLOv5, the construction site is monitored in real-time. Timely warnings are issued for those entering dangerous area to avoid accidents. Verified by the actual data, the method is timely and effective with high identification accuracy.