摘要
传统的人员安全行为识别方法存在准确率较低问题。为解决此问题,提出了基于YOLOv5算法的热电厂人员安全行为识别方法,对人员安全行为进行分类和标注,构建了适用于热电厂场景的数据集,通过采用特征提取和匹配算法,结合YOLOv5算法,构建了一个多目标检测模型,能够可靠地检测和识别各类安全行为。通过实验验证,此方法在热电厂场景中能够有效识别出人员安全行为,为人员安全管理提供可靠的辅助手段。
The accuracy of traditional personnel safety behavior identification method is low.To solve this problem,the study proposes a method for personnel safety behavior identification in thermal power plant based on YOLOv5 algorithm,classifies and labels personnel safety behaviors,constructs data sets which is suitable for thermal power plant scenarios and a multi-target detection model by feature extraction and matching algorithms combined with YOLOv5 algorithm.It can reliably detect and identify all kinds of security behaviors.The experimental results show that this method can effectively identify personnel safety behavior in the thermal power plant scene,and provide reliable auxiliary means for personnel safety management.
作者
赵文聪
Zhao Wencong(Beijing Gaojing Thermoelectric Branch,Datang International Power Generation Co.,LTD.,Beijing 100041,China)
出处
《黑龙江科学》
2024年第8期98-101,共4页
Heilongjiang Science