汽车疲劳驾驶的检测与监视系统设计  

Design of Driving Fatigue Detection and Monitoring System

作　　者：蔡木生[1] 周航 Cai Musheng;Zhou Hang(South China Institute of Software Engineering,Guangzhou University,Guangzhou 510990,China)

机构地区：[1]广州大学华软软件学院计算机系,广州510990

出　　处：《单片机与嵌入式系统应用》2020年第6期63-67,共5页Microcontrollers & Embedded Systems

摘　　要：针对疲劳驾驶易引发交通事故这一问题,本文设计开发了一个疲劳驾驶检测系统。该系统由树莓派、摄像头、轻量级云应用服务和移动端构成,实时获取驾驶员脸部视频,利用Dlib开源库得到人脸的68个关键点,通过计算其眼睛、嘴巴张合度,并与原来训练好的数据进行对比,依此判断驾驶员是否处于疲劳状态,若是,则进行语音提示,相关信息也能通过互联网上传到云端服务器中,车主或公司可通过移动端的微信小程序动态监控驾驶员状态。该系统具有成本低、易实现的特点,经测试,在有模型数据的情况下疲劳识别准确率为95%,可稳定运行。Aiming at the problem that fatigue driving is easy to cause traffic accidents,a driving fatigue detection system is designed and developed.The system is composed of raspberry pi,camera,lightweight cloud application service and mobile terminal.It can get the driver's face video in real time,and get 68 key points of the face by using Dlib open source library.By calculating the opening and closing degree of the eyes and mouth,and comparing with the original trained data,it can judge whether the driver is in fatigue state.If so,it can give voice warning,the relevant information can also be uploaded to the cloud server through the Internet,and the owner or company can dynamically monitor the driver status through the Wechat applet on the mobile terminal.The system has the characteristics of low cost and easy to realize.After testing,the accuracy of fatigue identification is 95%with model data,and it can run stably.

关 键 词：疲劳驾驶检测系统 图像处理算法 人脸检测 微信小程序 

分 类 号：TP51[自动化与计算机技术]