基于人眼开合度的电力调度员疲劳监测方法  

Fatigue monitoring method of power dispatcher based on eye opening and closing degree

作　　者：程正 杨帆 CHENG Zheng;YANG Fan(Guangdong power grid limited liability company Jiangmen power supply bureau,Jiangmen,Guangdong,529000,China)

机构地区：[1]广东电网有限责任公司江门供电局,广东江门529000

出　　处：《长江信息通信》2024年第3期46-48,共3页Changjiang Information & Communications

摘　　要：电力调度员参与电力系统运行的监视、指挥与控制的相关工作,是保证电力系统安全运行的重要成员,但长时间进行高度专注工作会引起疲劳,疲劳工作极有可能会导致严重事故。因此,开发一种基于人眼状态检测的疲劳监测系统尤为重要,有利于减少电力调度员因疲劳工作带来的安全和经济的损失。该课题设计一种基于YOLOv5和PERCLOS的组合模型的疲劳监测方法,模型由两个主要部分组成:人眼开合检测、疲劳状态判断。YOLOv5模型通过分析人眼开合度判断眼睛的眨眼状态,并利用PERCLOS算法检测人体疲劳状态。实验数据通过采用固定摄像头和可穿戴摄像头拍摄来获取,通过对比分析说明可穿戴摄像头拍摄的数据更可靠,同时通过YOLOv5模型与dlib模型的对照实验可证明YOLOv5模型具有识别准确率高的特点。Power dispatchers participate in the monitoring,command and control of power system operation and arc important members of ensuring the safe operation of the power system,but long-term highly focused work will cause fatigue,and fatigue work is likely to lead to serious accidents.Therefore,it is particularly important to develop a fatigue monitoring system based on human eye condition detection,which is conducive to reducing the safety and economic losses caused by fatigue work of power dispatchers.In the projct,a fatigue monitoring method based on the combined model of YOLOv5 and PERCLOS is designed,which consists of two main parts:human eye opening and closing detection and fatigue state judgment,YOLOv5 model determines the blinking state of the eye by analyzing the opening and closing degrce of the human eye,and uses the PERCLOS algorithm to detect the fatigue state of the human body.The experimental data were obtained by using fixed camera and wearable camera,and the data taken by wearable camera was more reliable through comparative analysis,and the control experiment betwcen YOLOv5 model and dlib model could prove that the YOLOv5 model had the characteristics of highaccuracy.

关 键 词：疲劳监测 YOLOv5模型 人眼开合检测 疲劳状态 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]