Drowsy Driver Monitor System Research And Design

Driving in case of drowse is a most important factor of traffic accidents. Now many countries are active in studying drowsy monitor. The article presents an infrared collection system. In this system, according to the study of physiological changes about the fatigue, eye changing information is used to detect the degree of fatigue. Not affecting the driver, the system can detect fatigue by means of machine version and image processing analysis.A math model of the combination of PERCLOS and blinking times is used to compute the degree of fatigue after studying on much physiological information.Light takes great effect on image collection, so the system uses infrared illuminator in order to work in full time. In addition, a 950nm infrared filter is used to eliminate visible light and infrared light outside the 950nm narrowband. These can improve the improve image definition. After compared many methods, the article presents a combination of Matlab and 1394 video camera.We know the pupil's total area, eye closing degree and blinking times by means of computing pupil area in real time. So the mission of the image processing is to locate the face and eyes, and then calculate the area of pupils that are not being covered by the eyelids. The image processing analysis is divided into face location and eyes location. In processing of face location, difference image method and optical flow analysis is used to detect the movement area. After that we use class deviation maximization segmentation method and other assistant methods to detect the max movement area and consider the area as the face location.In processing of eye location and pupil area computing, we use circle filter to locate the middle of line of the two eye centers and then estimate the eye area, locate eye center, compute the visible pupil area. In order to improve efficiency, eye location of the next fame is predicted according to previous ones.Many tests show that the system can detect blinking times and visible pupil area in real time. So it can detect the degree of fatigue efficaciously.