According to the relevant investigation information of the Ministry of communications of China, the three main factors inducing traffic accidents are drunk driving, speeding and fatigue driving. Among them, for drunk driving and speeding driving, our country has carried on the regulation and governance from the legal level, but there is still a lack of effective prevention and governance measures for driver fatigue driving.

When the driver continues to drive for a long time without getting enough rest, his physiological and psychological functions will gradually be out of balance. His ability to control the vehicle, his attention and his ability to respond to emergencies will be significantly reduced, and he may even fall into a short sleep, leading to the complete loss of control of the vehicle. This phenomenon is called fatigue driving.

Up to now, the design and development methods of fatigue driving detection system at home and abroad can be divided into subjective detection method and objective detection method. Subjective detection method is to obtain the driver’s psychological, driving action and facial expression information in different time periods through manual inquiry, questionnaire filling and subjective evaluation, and then obtain the driver’s fatigue state through the analysis of the obtained information.
Different from the subjective detection method, the objective detection method can avoid human interference, but mainly through the external device to collect the relevant data in the driving process, and then rely on the advanced artificial intelligence algorithm to analyze and evaluate whether the driver is in the fatigue driving state. Objective detection method will not be affected by the driver’s subjective consciousness, so it has higher accuracy and better real-time performance. Because of the advantages of objective detection method, it has become the main research direction of driver fatigue detection. Objective detection methods can be subdivided into four types

1. Paper Inverstor for Fitigue Driving

The detection method based on motor vehicle behavior characteristics, when the driver presents fatigue state, to a certain extent, it will be reflected by the vehicle behavior he operates. These vehicle behavior characteristic parameters can be easily obtained and used for fatigue detection, such as the pressure from the driver’s hand on the steering wheel, the change of vehicle acceleration, the rolling of the body and the vehicle By measuring the corresponding data of these vehicle behavior characteristics in the driver’s normal driving state and fatigue driving state, we can formulate a judgment standard to distinguish the normal driving state and fatigue driving state, and then compare the real-time measurement data with this standard, so as to analyze and judge the driver’s fatigue state. The disadvantage of this kind of detection method is that it is easily affected by the natural environment, the driver’s driving level, psychological quality and driving mood.

2. Body Sensor Detection for Fitigue Driving

The physiological characteristics of drivers mainly refer to ECG, EEG, body surface temperature and so on. Generally, they need the help of medical devices and related equipment to complete the acquisition work. Similar to the detection method based on motor vehicle behavior characteristics, it needs to collect the physiological characteristics data of the driver during normal driving and fatigue driving, generate the corresponding driving state differentiation standard, and then compare the real-time data collected by the driver in the driving stage with the two data standards to judge the driver’s driving state. This kind of method is a kind of contact method The detection method is simple.
Although the detection method based on the driver’s physiological characteristics has high accuracy, the driver needs to wear relevant physiological index detection equipment, which will affect the comfort in the driving process and bring inconvenience to the driver.

3. Faceial DMS for Fitigue Driving

Detection method based on driver’s facial features detection method based on driver’s facial features is to extract and analyze driver’s facial features such as eyes open and close, mouth open and close, and head posture based on the driver’s facial images obtained by hd camera sensors such as cameras, etc.  Analysis, so as to realize the analysis and judgment of the driver’s fatigue state. The detection method based on the driver’s facial features is low cost, no need to touch, and easy to detect, so it has become a hot research direction of driver fatigue detection system.
DMS-Drivier Fitigue Driving Monitor