EEG Signal Driving Fatigue Detection based on Differential Entropy

With the gradual increase of vehicle penetration rate, driving safety has become a problem that people pay attention to, and traffic accidents caused by fatigue are mostly, therefore, it is meaningful to devote ourselves to research on the method of detecting driving fatigue so as to achieve safety...

Full description

Saved in:
Bibliographic Details
Published inInternational Conference on Industrial Mechatronics and Automation (Online) pp. 543 - 548
Main Authors Wang, Danyang, Tong, Jigang, Yang, Sen, Chang, Yinghui, Du, Shengzhi
Format Conference Proceeding
LanguageEnglish
Published IEEE 04.08.2024
Subjects
Differential Entropy
Driving Fatigue
EEG
Electroencephalography
Entropy
Fatigue
Feature extraction
Gradient Boosting Decision Tree
Mechatronics
Nearest neighbor methods
Support vector machines
Online AccessGet full text

Cover

More Information
Summary:With the gradual increase of vehicle penetration rate, driving safety has become a problem that people pay attention to, and traffic accidents caused by fatigue are mostly, therefore, it is meaningful to devote ourselves to research on the method of detecting driving fatigue so as to achieve safety warning. Based on this, in this paper, we collect the EEG signals of the experimenter through driving simulation, then extract the features through differential entropy, classify the features with GBDT, and finally determine the fatigue level of the subject with the classification results. We use our method to analyse the difference with sample entropy and fuzzy entropy at the feature level, and with KNN and SVM at the classification level. The comparison shows that our method has significant accuracy in EEG-based fatigue detection.DE can be used accurately in extraction methods through its properties, and differential entropy-based GBDT methods may be useful in detecting driver fatigue.
ISSN:2152-744X
DOI:10.1109/ICMA61710.2024.10632910