| Nowadays,the automobile plays an important role in people's life.There is no doubt that it brings great convenience to people,but at the same time,it also produces a huge security risk—— fatigue driving.In-depth analysis of the causes of major traffic accidents at home and abroad,a considerable part of the reasons is caused by driver fatigue.In order to avoid this kind of traffic accidents and respond to the call of the strong police of science and technology,it has become an important subject of practical significance to study an efficient and reliable fatigue driving early warning system.At present,the detection of fatigue driving is mostly based on vehicle trajectory,driver behavior or fatigue judgment from a single physiological index,which have some defects.These detections are vulnerable to the impact of road environment and driving habits of drivers,and the accuracy is low.Moreover many researches on fatigue driving,such as EEG,ECG and EMG,need to paste some equipment on the driver,which makes the discomfort to the driver extremely intolerable.Although this kind of contact detection can obtain accurate physiological data of drivers,but in terms of the whole monitoring process,it also affects the driver's driving and normal driving fatigue performance.The noncontact detection overcomes these effects on the driver.The traditional noncontact detection can judge the fatigue by capturing the blink state of the human eye or the driving characteristics of the vehicle.The poor real-time performance and the single detection index result in the low reliability of the system.In order to overcome the above problems,these paper constructs a fatigue driving detection system based on multi-source information fusion of artificial neural network from the perspective of fatigue generation mechanism and fatigue symptoms,taking into account a variety of characteristic signal indicators.The main research contents of this paper are as follows:(1)The research status of fatigue driving at home and abroad is analyzed,and the method of fatigue driving detection based on driver's eye characteristic signal,yawning signal and ECG signal is proposed,and these signal indexes are described in detail.(2)A simulation driving platform is built in the laboratory,a set of clear and rigorous experimental paradigm is designed,and the information of driver's eye and mouth is collected through the camera,the driver's face is located and tracked through the camera,and the driver's eye features and yawning state are detected,and the driver's ECG signal is detected through the intelligent bracelet,providing data support for the final algorithm.(3)The fatigue threshold is obtained by processing and analyzing the data of each characteristic index,and the specific fusion algorithm is used to form the comprehensive fatigue judgment index to detect the driving state of the driver during the driving process and identify the fatigue.Experiments show that the system can effectively detect the driver's fatigue state and give an early warning.The multi-source information fusion ensures the reliability and efficiency of the system.This paper is the main part of the Cerent project ‘Fatigue Driving Detection and Early Warning Platform Based on IPv6'(NGII20170712).This paper focuses on the detection of fatigue driving based on the fusion of facial multi features and physiological signals.This is not only a frontier basic scientific issue,but also an important guarantee for the safety of people's lives and property. |
PDF Full Text Request