Driver Fatigue Detection System Based On The Human Eye Detection Research And Implementation

Fatigue driving is one of important reasons which is leading to traffic accidents. In thedriver factors causing traffic accidents, fatigue driving is one of the primary factors whichresult in death. The incidence of traffic accidents induced by fatigue driving also increases inChina every year as the number of automobile is increasing rapidly, fatigue driving hasbecome a major threat to the personal safety. Therefore, fatigue detection system has greatsignificance to assure safety of people.Widely absorbed some new advantages of driving fatigue detection methods of thedomestic and abroad, lots of researches has been done in this thesis. A driver fatigue detectionsystem is implemented based on face detection and eye detection.Finally, the proposedmethod was testified by experiment. The main work in this thesis is listed as follows:(1)Domestic and international research of driving fatigue detection system is summarizedand forecasted, this thesis issues the software system and hardware device for driving fatiguedetection system.(2)Based on existing face detection method，a method for the face detection based on skincolor model is feasible. In order to improve the accuracy and speed of face detection, thisthesis proposes an improved detection scheme based on the facial feature. The experimentalresult shows that this method of face detection has a good performance in complexbackground scene.(3) A new method of eye location based on light reflection characteristics is developed inorder to ensure the accuracy in the complicated circumstances. This method has low quantityof calculating and high accuracy, it can meet the needs of real time processing.(4)Calculating blinking frequency by PERCLOS is to determine the driver fatigue’sdegree. The process of driver fatigue detection can be accomplished in this way.