Driver fatigue detection from electroencephalogram spectrum after electrooculography artefact removal. Issue 1 (1st March 2013)

• Record Type: Journal Article
• Authors: Hu, Shuyan; Zheng, Gangtie; Peters, Björn
• Is Part Of: IET intelligent transport systems. Volume 7:Issue 1(2013)
• Page Start: 105
• Page End: 113
• Publication Date: 2013-03-01
• Subjects: adaptive filters -- driver information systems -- electroencephalography -- feature extraction -- independent component analysis -- road safety -- support vector machines; driver fatigue detection -- electroencephalogram spectrum -- electrooculography artefact removal -- driver fatigue monitoring -- detection system -- fatigue‐related traffic accidents -- drowsiness detection -- EEG power spectrum analysis -- independent component analysis with reference -- electrooculography artefacts removal -- proposed ICA‐R algorithm -- adaptive filter -- EEG spectrum feature extraction -- EEG spectrum‐related features -- support vector machine recursive feature elimination; Intelligent transportation systems -- Periodicals; Electronics in transportation -- Periodicals; 388.31205
• DOI: 10.1049/iet-its.2012.0045
• Languages: English
• ISSNs: 1751-956X

Abstract

Abstract : A driver fatigue monitoring and detection system with high accuracy could be a valuable countermeasure to decrease fatigue‐related traffic accidents. This study proposes methods for drowsiness detection based on electroencephalogram (EEG) power spectrum analysis. First, a new algorithm is proposed for independent component analysis with reference (ICA‐R) for electrooculography artefacts removal. Comparison is then carried out between the proposed ICA‐R algorithm and an adaptive filter. Secondly, 75 EEG spectrum features are extracted from the cleaned EEG. Among all the EEG spectrum‐related features, 40 key features are selected by support vector machine recursive feature elimination to improve the performance of the classifier. The validation results show that 86% of the driver's drowsiness states can be accurately detected among drivers, who participate a driving simulator study.