基于云平台的疲劳驾驶检测系统  

Fatigue Driving Detection System Based on Cloud Platform

作　　者：张廷[1] 熊章锦 周小凯 谢谦 ZHANG Ting;XIONG Zhangjin;ZHOU Xiaokai;XIE Qian(School of Information Engineering,Minzu University of China,Beijing 100081,China)

机构地区：[1]中央民族大学信息工程学院,北京10081

出　　处：《中央民族大学学报（自然科学版）》2020年第3期61-66,70,共7页Journal of Minzu University of China(Natural Sciences Edition)

基　　金：北京市大学生创新项目(2019Beij110021)。

摘　　要：本文针对国内因疲劳驾驶而事故多发的情况,提出了基于深度机器学习和智能云平台的疲劳检测和驾驶事故自动检测解决方法。该方法很好地适应了车辆和驾驶室的复杂工作环境,旨在让操作人员能够快速实现基于深度和机器视觉的疲劳驾驶检测和报警功能,通过在驾驶员的上方安装可移动摄像头实时采集驾驶员的图像,将实时获取的视频流数据送入处理设备进行处理,同时利用预先训练好的疲劳检测模型从摄像头提取出人脸的特征点,依据PERCLOS疲劳驾驶判断准则来判断车辆驾驶者是否已经处于疲劳状态并实时上传图像数据至云端处理平台,同时根据人脸的位移和方向,利用舵机使摄像头始终正对驾驶者的脸部,最终完成一个集便携性、高效性、准确性为一体的疲劳检测系统。In order to effectively deal with the complex situation of fatigue detection and frequent accidents of driving,a solution of automatic fatigue detection of driving based on deep machine learning and intelligent cloud platform is proposed. The solution is well adapted to the complex driving environment. The purpose is to help the operator to quickly realize the fatigue driving detection and alarm function based on depth and machine vision,and collect the driver’s image by a movable camera,The real-time video stream data is sent to the processing equipment for corresponding processing. At the same time,face feature points are extracted by using the pre-trained fatigue detection model to monitor the dynamic changes of the eyes and mouth of the driver. According to PERCLOS fatigue driving judgment criterion,we can judge whether the driver is in fatigue driving state and upload image data to the fatigue driving cloud processing platform. Finally,a fatigue detection system with portability,high efficiency and accuracy is completed.

关 键 词：疲劳驾驶 疲劳检测 物联网 人脸识别 

分 类 号：U491.6[交通运输工程—交通运输规划与管理] TP391.4[交通运输工程—道路与铁道工程]