基于图像处理的疲劳驾驶预警系统设计与实现

广东东软学院本科毕业设计（论文）ABSTRACTWith the continuous development of science,technology andeconomy,people's living standards are also greatly improved.However,the popularity of motor vehicles has brought a high traffic accident rate.According to the investigation of relevant organizations,nearly one fifthof traffic accidents are caused by drivers'fatigue driving.Therefore,weshould pay attention to the severity of fatigue driving.When the driverdoes not rest for a long time,peoper will have blurred vision,inflexiblemovement and lack of concentration.When the driver is tired driving,itcan be divided into three situations,mild,moderate and severe.Thesesituations will lead to the occurrence of traffic accidents.Therefore,inorder to reduce the occurrence of traffic accidents in our country,weshould start with fatigue driving.Although our country's road traffic lawclearly defines fatigue driving and issues punishment regulations,butfatigue driving is still prohibited for many times,so it is very important toprevent drivers and passengers from fatigue driving to carry out on-boarddetection.Now,the application of fatigue driving early warning system inthe market has not been popularized,and more factors are the problem ofaccuracy and low cost.In this paper,the problem of recognition rate and cost is studied.Firstly,the research direction and purpose of this paper are defined,and