Condition Monitoring Techniques Driving

With the growth of social and economic, the rising of car ownership, road traffic accidents have become a major threat to people’s lives and safety, the driver fatigue is one of the main reasons of the accident, so the real-time monitoring and prediction of driver fatigue is a very important research direction.This paper designed and completed a monitoring and warning system for driver fatigue based on BF533and PIC microcontroller. The system through a camera monitoring the driver, judge the driver fatigue through the mixed multiple facial feature (including the eyes and mouth movements). If it is true, the microcontroller will get the warning system to start. System used the image difference to locate the eyes by the "red-eye" effect, and then locate the mouth by the distribution of features. After the location of the necessary clues, the system used template matching for eye feature extraction as well as the binarization and Gabor wavelet for the mouth. Through post-judgment PERCLOS and L D A, using Bayesian network mix the single feature, to obtain the final fatigue judgment.The positioning method is simpler the regular way, reduced the amount of computation, improved work efficiency; adopt multi-feature fusion method also makes monitoring results more accurate, more robust; have done some optimization for the warning system, introduced the velocity to adjust the Sensitivity, and add Bluetooth Car massage, relieve fatigue at the same time warning the driver.