Research On Fatigue Driving Detection Technology Based On EEG Signal Analysis

Fatigue driving has gradually become the key cause of traffic accidents.If relevant technologies can be used to remind drivers of fatigue driving,the probability of traffic accidents can be greatly reduced.As a key signal of individual state change,eeg can provide technical support for driver fatigue state detection.This thesis studies the fatigue driving detection technology based on EEG signal acquisition.The main work of the thesis is as follows:(1)In order to solve the problems of difficulty in distinguishing fatigue states and signal nonlinearity in the process of EEG signal acquisition,this thesis distinguishes and extracts the features of EEG signals based on EEG signal entropy reconstruction and wavelet packet decomposition technology.In order to ensure the real-time acquisition of EEG signals,blink frequency,power spectral density and mental concentration were selected as variables to comprehensively analyze the driver’s fatigue quantification method.The effect of power spectral density on the driver’s state discrimination is analyzed,and the correlation between the driver’s mental concentration and fatigue detection is proved.(2)In order to solve the problem of lack of identification methods in the process of EEG signal detection.In this thesis,a driver state recognition method is designed combined with deep neural network,and the overall framework of recognition is established.This detection method proposes a direct classification and recognition functional brain that can be combined with the spatial domain convolution method of B-spline curves.By comparing the various characteristics of this method with the traditional method,it is found that the functional brain network fatigue driving detection method designed this time is better than the traditional detection method in all aspects.(3)In order to solve the problem of signal instability caused by various disturbances in the process of EEG signal processing,this thesis analyzes the performance of fixed mode and artifact denoising.And combined with the experimental data of brain wave data acquisition,the stability characteristics of this graph neural network fatigue driving were analyzed.The research method is analyzed through simulation and control experiments,and the experimental data is analyzed.The results show that the functional brain network fatigue driving detection method is better than the traditional detection method in all aspects.The fatigue driving detection method has stability in detecting the physiological fatigue condition of driving volunteers.It has certain practical significance for the fatigue driving detection technology.