Study On The Methods For Driver Fatigue Detection Based On Video Analysis

Statistics show that driver fatigue is an important factor in the increasing number of accidents on today's roads. Many countries make great efforts on how to detect fatigue during driving. Therefore, how to effectively monitor and prevent driver fatigue driving has much real significance to reduce traffic accidents and personnel mortality.The purpose of this project is to advance a system to detect fatigue symptoms in drivers. In this paper, the video analysis method for driver drowsiness detection is studied. The position of driver's face is obtained based on the analysis of driver video. This paper describes how to find the eyes and mouth. Eyes and mouth cues relating to drowsiness are extracted based on position of the face. By analyzing the status of the eyes and mouth determined whether driver is tired.The main work done in this thesis is as following:(1) Locatoin of driver's face. In order to improve veracity we do some necessary pretreatments before face location, such as illumination intensity emendation and removing noises.The system uses skin color information and blob analysis for detecting the face pixels of driver. The skin color has the good assembling character in the color room. This thesis selects YCbCr color room as distributed and statistical mapping room, and builds a two-dimensional Gause distributed mathematical model through statistics and recognition. It adopts face detecting method based on similarity and the shape of face.(2) Location and state analysis of driver's eyes. In order to improve efficiency we do some necessary pretreatments before eye detection, such as changing RGB images to gray level images and gray expansion. We use a new differential and integral combination projection to locate the two eyes. Determined whether the eye is open or closed by projection curve.Eye is tracked by tracking eyebrows used Kalman filter.(3) Location and state analysis of driver's mouth. Lips color is reder than other face skin color. Based on this point, we select rgb color room which is transformed from RGB color room. This paper presents a new method to detect the driver's lips. The foundation of this method is that the ratio of the three color elements of two points is equal if they have same color. Then we obtained binary images of the driver's lips from above analysis. Next make project used an ordinary integral projection in the vertical and horizontal direction to the binary images of the driver's lips. According to the projection curve we work out the width and height of the lips.The ratio of width and height is calculated for judging whether driver is yawning.This paper adopts MATLAB7.0 to develop the algorithm of the location of the driver's face in the driver fatigue monitor system, and makes the related test to show the correctness of this study.