Based On Detecting Driver's Fatigue Status Of Vehicle Active Safety

With the development of society, motor driver fatigue is increasingly reported as a problem concerning the road traffic safety. Recent research reveals that approximately 25-30% of vehicle collisions result from driver fatigue. The chance for fatalities is very high.Many of us have experienced sleepiness while driving. Fatigue may adversely affect drivers'alertness and ability to drive safe. The study situation on detecting and evaluating techniques of motor driver fatigue in the world is reviewed in this paper. But there is not a good system making traffic ancient cutting down. An active safety system with human center would be cut down traffic ancient. This is an important field of ITS. Research about fatigue detection system of drivers in the world is introduced at first, and active safety system with human center is introduced. A basic idea is set up: based on computer vision, image procession, mode distinguish, analyses active security of drivers, an active safety system with human center would be set up finally.An advanced snake algorithm and an advanced Hough Transforms algorithm be used in image procession and mode distinguish field; outline and character of face are abstract. Especially, an important problem of location rotating face has been solved in this paper. Fatigue state of drivers is detected by PERCLOS algorithm.Character of information in motor moving and three period of driving processing are discussed in this paper: senor, making decision, operation. In close system, there is relation of human-motor-environment. A model of relation of human-motor- environment is a center of human.Based on conventional model of making decision, aim at characters of driving procession, a model of based on self-learning making decision is raised. Aim at every link of model, based on evidence merge character match, based on rules logics inference, based on character conformity coherence analyses, based on judgment of security choosing procession model, based on projection transforms self-learning model.According to affect of security sensor in driving procession, measure of security sensor as a new concept has been raised. Based on measure of security sensor statistic model, security degree of human-motor-environment has been raised. Security sensor of drivers is judged by statistic model. According to security sensor of drivers in fatigue state, based on security sensor fatigue analysis algorithm has been raised.Based on acquire information, according to anglicizing several classical moving state, based on minim security distance dangerous estimation has been raised.