Low Delay And High Precision Embedded Remote Monitoring System For Fatigue Driving

As the number of motor vehicles in the country increases year by year,people’s travel is increasingly inseparable from cars.While we enjoy the convenience that vehicles bring to our daily life,it is also accompanied by the occurrence of dangerous accidents.In recent years,the incidence of traffic accidents has increased year by year.Fatigue driving has become one of the main causes of traffic accidents.Therefore,it is urgent to launch early warning and protection against fatigue driving.Most of the equipment for fatigue driving detection in the domestic market is cumbersome and complicated to assemble,and it is impossible to obtain the relevant information of the driver in the car in the first time,and it is impossible to reduce the risk of accident in time.In view of the above situation,this article discusses the design of an embedded remote monitoring system for fatigue driving,and specifically introduces the design process of an embedded lightweight fatigue driving monitoring system based on the RTMP transmission protocol.The main research contents include:designing a lightweight SSD network structure to optimize the performance of embedded face detection,and combining Dlib face key point detection algorithm,PERCLOS fatigue judgment standard to realize the detection of fatigue driving behavior;and through analysis The development status of the video surveillance system puts forward a remote real-time video surveillance system based on RTMP+Nginx,which reduces the delay problem in the video transmission process by reducing GOP and multi-threading.Finally,the fatigue driving detection algorithm and the streaming media server are deployed on the same embedded development board to achieve a low-latency and high-precision embedded fatigue driving remote monitoring system.The staff can realize remote monitoring through any streaming media player,the first time Grasp the information in the car to reduce and prevent accident risks in a timely manner.Experimental results show that the system has a recognition rate of 95.2%for fatigue driving behavior,and the long-distance video transmission delay is about 600 milliseconds,which meets the needs of embedded fatigue driving monitoring and real-time video transmission.