Research And Application Of Fatigue Monitoring Technology Based On EEG

Mental fatigue is a subjective discomfort caused by prolonged concentration or other mental activity.Mental fatigue can affect the efficiency of learning and work,and can even lead to accidents,resulting in the loss of personnel property.Electroencephalogram(EEG)records changes in brain electrical activity as a general reflection of the electrophysiological activity of the cerebral cortex or cerebral neurons on the surface of the scalp.When the brain is in a state of fatigue,it is often possible to use the EEG signal to determine if the brain is fatigued.The EEG signal is an unsteady and nonlinear biosignal that makes this signal processing more difficult.In recent years,the monitoring of mental fatigue status based on EEG signals has gradually attracted the attention of scholars in the field of EEG.The study of EEG signals in mental fatigue has become a hot topic in the field of EEG signals.Based on the previous studies,this paper mainly completes the following work:(1)The experiment of designing and completing the EEG signal in the state of mental fatigue is different from the brain fatigue induction experiment of the single reference dimension.This paper uses the psychology experiment stroop experiment to design the fatigue state of the human brain induced by two different difficulty levels.During the experiment,EEG signals were collected under different degrees of mental fatigue.(2)The original EEG signal is preprocessed by wavelet threshold function denoising based on integrated empirical mode decomposition.This paper uses the optimization of traditional wavelet threshold function to overcome the soft threshold and hard threshold defect.The integrated empirical model used in this paper State decomposition is an improvement of the traditional empirical mode decomposition,which overcomes the deficiencies of the traditional empirical mode decomposition method.Combining the two,the wavelet threshold denoising method is used to process the highfrequency signal to solve the shortcomings of the integrated empirical mode decomposition for the low-frequency signal processing.Experiments show that compared with the traditional denoising method,wavelet threshold denoising based on integrated empirical mode decomposition can better applicability and practical significance to mental fatigue EEG signals.(3)In view of the nonlinearity and non-stationarity of brain fatigue EEG signals,this paper analyzes the subjective perception of mental fatigue state from the energy and complexity,and analyzes the energy entropy,sample entropy and fuzzy entropy of EEG signals.The correlation between mental fatigue.The multi-dimensional feature vector proposed in this paper combines the energy feature and the nonlinear feature with respect to the single energy entropy as the feature vector.According to the classification result obtained by the input support vector machine,the recognition rate of the brain fatigue signal for mental fatigue is about 20 % increase.