The Study Of Fatigue Driving Detection Method Based On Monocular Vision

With the rapid development of economy, people's living standards have been significantly improved and the number of private cars is also increasing. At the same time, the traffic accidents are more and more because of various reasons. Among these factors which caused traffic accidents, the traffic accidents caused by driver fatigue are 80%. In order to keep the driver awake during driving, and create a safe traffic environment, we take an effective fatigue driving detection method to monitor the driver's fatigue level real-time, and alarming when the driver will be tired, which has important practical significance for personal security and the development of social.Traditional fatigue driving detection method has the problems of low detection accuracy, time-consuming. In order to improve the accuracy and real-time of the detection method, a new method based on monocular vision is proposed in this paper. Collecting the driver's head image through a single visual image sensor, in order to eliminate the effects of different light, using the homomorphic filter for facial Image; Using the histogram equalization processing to enhance the contrast of the image and improve the accuracy of face detection and feature extraction. In the process of face detection, the face detection algorithm based on AdaBoost is improved, and the improved face detection algorithm is used for face detection; Using Feature extraction algorithm based on grayscale integral projection to extract features of the eyes and mouth in the detected face region, locating the coordinates of the mouth and eyes; According to the principle of projection transformation, the head pose is calculated with the feature points of the face; Finally, the head position and orientation are calculated according to the head posture, and the degree of eye closure and the blink frequency are calculated according to the feature of eyes, which are used for fatigue detection.Designing simulation experiments by simulating the driver's driving state, detecting the face image under different illumination conditions. Analyzing the detection accuracy and the speed of Face detection algorithm, eye and mouth detection algorithm, head pose estimation algorithm based on facial feature points, and analyzing comprehensively the accuracy and speed of fatigue driving detection algorithm based on head posture and eye features. Lots of experiments show that using the detection method in this paper to detect fatigue under different illumination conditions, which can keep high accuracy and short detection time, and can meet the requirements for real time and accuracy of fatigue detection.