基于YOLOv8-pose的施工人员疲劳姿势监测和调整系统研究  

Study on Fatigue Posture Monitoring and Adjustment System for Construction Workers Based on YOLOv8-pose

作　　者：李斌 LI Bin(China Railway No.18 Bureau Group No.1 Engineering Co.,Ltd.,Baoding 072750,Hebei,China)

机构地区：[1]中铁十八局集团第一工程有限公司,河北保定072750

出　　处：《建筑施工》2025年第4期576-581,共6页Building Construction

基　　金：福建省住房和城乡建设厅科学技术计划项目(2023-K-31,2023-K-63,2023-K-73)。

摘　　要：量化施工人员疲劳程度对减少施工不安全行为、保护工人生理性健康和提高现场的安全管理水平具有重要意义。基于OWAS和RULA姿态评估方法建立疲劳评价指标,并依托NOKOV动作捕捉系统和YOLOv8-pose算法检测分析施工人员的疲劳等级,建立了施工人员疲劳作业姿势评估和方案调整系统。通过试验验证了所提系统的有效性和可行性,结果表明:所提方法能够准确反映现场施工人员的疲劳程度,所提系统能够实现施工人员疲劳程度的自动化检测并提供对应的调整方案。Quantifying the fatigue level of construction workers is of great significance in reducing unsafe construction behaviors,protecting workers'physiological health,and improving safety management on site.A fatigue evaluation index is established based on OWAS and RULA posture evaluation methods.The fatigue level of construction workers is detected and analyzed using the NOKOV motion capture system and YOLOv8-pose algorithm,and a system for evaluating fatigue posture and adjusting the scheme is developed.The effectiveness and feasibility of the proposed system are verified through experiments.The results show that the proposed method can accurately reflect the fatigue level of construction workers,and the system can automatically detect the fatigue degree of construction workers and provide corresponding adjustment recommendations.

关 键 词：疲劳监测 施工人员 深度学习 计算机视觉 作业姿势评价 

分 类 号：TU741[建筑科学—建筑技术科学]