Fatigue monitoring method of power dispatcher based on eye opening and closing degree
Power dispatchers participate in the monitoring,command and control of power sys-tem operation and are 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 seri-ous 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 project,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 mod-el determines the blinking state of the eye by analyzing the opening and closing degree 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 between YOLOv5 model and dlib model could prove that the YOLOv5 model had the charac-teristics of high accuracy.
fatigue monitoringYOLOv5 modelhuman eye opening and closing detectionfa-tigue state
NETL
NSTL
万方数据
