Design And Research On Real-time Fatigue Driving Detection System For Driver

With the quantity of automobile and the road rank continue increasing, the world road accident occurs frequently, traffic safety environment is worse day by day. An investigation showed that the driver fatigue driving is a major cause for fatal traffic accidents. Therefore, it has great significance and application value to study and design an accurate and efficient driver fatigue detection system to warn the drowsy driving in real-time.Because of the advantages of non-contact, real-time, accuracy and so on, the method of fatigue detection based on the state of facial features has become the main research direction for fatigue detection. Combined with funded projects of “Twelfth Five-Year Plan” of the Education Department of Jilin Province(No. 2015097) and Development and Reform Commission of Jilin Province(No. 2015Y067), a driver fatigue detection system based on the driver’s facial feature has been established in this article after the fatigue driving detection technology is summarized at home and abroad. In this system, many techniques, such as machine vision, image processing, target detection and feature extraction, are used to detect and analyze the fatigue-related external physiological feature, and then the related fatigue criteria is used to judge fatigue state. Four modules are mainly included in this system, which are image capture module, image preprocessing module, feature detection and extraction module and feature classification recognition module. And modules of feature detection and classification recognition are studied emphatically, which mainly include the driver’s face detection, face tracking, facial features detection and state recognition, as well as fatigue recognition based on facial features and analysis.The main research contents of this article are as follows:Firstly, a detection system platform is built by using computer, Visual Studio and some other software and hardware equipment. On the basis of image preprocessing, a face classifier is trained by selecting MP-LBP as the classification feature and using the method of Ada Boost algorithm and cascade, which can detect and locate the driver’s face region in real-time. Aiming at the shortcoming of Camshift tracking algorithm, an algorithm combining the Camshift tracking algorithm with Kalman filter is proposed to realize the realtime tracking of human face region.Secondly, the face model is obtained by training the sample images calibrated the facial feature points by using the gradient regression tree algorithm. The regions of eyes and mouth can be located by using this face model on the detected face. Then according to the related fatigue criteria, the degree of driver fatigue can be judged by analyzing the state of eyes and mouth which is determined by using the ellipse fitting algorithm and setting the threshold.Finally, to verify the accuracy of the proposed driver fatigue detection algorithm, a fatigue driving detection experiment is carried out in the Honda’s car. The driver’s face images are captured by installing the COMS camera with infrared function on the front windshield, and the data are calculated and analyzed by computer. Experiment contents include the face region detection and tracking, facial features detection and state recognition, as well as fatigue recognition based on facial features and analysis. The experiment results show that the system has good accuracy, real-time and robustness.