The Research On Driver Fatigue Detection Algorithm Based On Eye Detection

With the development of social economy and the improvement of people's life,communication and network application have been developing rapidly.For the construction and improvement of road traffic,cars have become one of the most important means of transport in people's daily work and life.By the end of 2015,China's car ownership has been the second in the world with the number of 172 million while American is the first.Smooth road,increasing number of cars and faster speed are main reasons for frequent traffic accidents,which causes irreparable damage to people's lives and property.what's more,fatigue driving is also an important cause for traffic accidents.In recent years,the pressure from life and work make drivers fatigue problems more and more severe.So the importance of driver fatigue detection is improving.So as to solving the problem of driver fatigue driving,many domestic and international experts and scholars put forward many fatigue detection methods,Such as the direct method based on physiological parameters and reflection,the indirect method based on the behavior of the operation and wheel path,and the method of image processing.Due to the restrictions of driving environment,some of the methods that need to contact the drivers directly to detect the physiological parameters and operational behavior will affect the driver's operation,leading to security problems.In addition,the indirect detection of fatigue by vehicles is easy to make a wrong judgment.Therefore,the driver fatigue detection based on image processing has attracted the attention of scholars.But the real-time,accuracy and operability of the system have a big gap with the actual application.This article based on the analysis of the driver fatigue detection algorithm,aiming at promoting the accuracy of fatigue detection method of image,discusses the key research on fatigue detection algorithm based on human eye.And the driver's color space face detection algorithm and the edge detection eye location algorithm are studied and improved.Finally,the validity of the method for driver fatigue detection is verified through experimental simulation.The main contents of this paper include:1.Analyze the research status and characteristic of driver fatigue detection algorithm;2.Introduce the knowledge and technology of image preprocessing;3.Discuss the common methods of face detection,and deeply analyze the color face detection algorithm based on skin color model.4.Deeply study the human eye detection of classical edge operator,and give a detection algorithm based on direction of Sobel operator.5.Establish the human eyes library,complete the fatigue state detection based on BP neural network,and provide the simulation results.