Please use this identifier to cite or link to this item:
Title: An eeg based study of unintentional sleep onset
Keywords: EEG, drowsy, driving, SVM, eyeblinks, alert
Issue Date: 20-Aug-2008
Citation: YEO VEE MIN, MERVYN (2008-08-20). An eeg based study of unintentional sleep onset. ScholarBank@NUS Repository.
Abstract: The research objective is to effectively detect driver drowsiness by (1) identifying characteristic EEG differences between voluntary recumbent and involuntary sleep onset, (2) establishing an automatic method of distinguishing between alert and drowsy states by Support Vector Machines (SVM), (3) constructing a Driver Drowsiness Index (DDI) to develop a reliable drowsiness detection system.Recumbent sleep tests, day and night driving simulations were conducted on thirty subjects. Alert and drowsy EEG data segments were marked and used to train binary and multi-class SVM tools using a distinguishing criterion of four frequency features across four principal frequency bands. Vertex sharpness was significantly sharper and triple conjoined vertex waves only occurred during voluntary recumbent sleep onset. A conjoined vertex spindle waveform was statistically associated with driving sleep onset. SVM binary-class classification between alertness and drowsiness achieved 99.3%. The DDI classified the sleep onset process into 5 levels, achieving 77.2% accuracy by SVM multi-class classification.
Appears in Collections:Ph.D Theses (Open)

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
PhD Thesis_Mervyn Yeo Vee Min_2008.pdf3.83 MBAdobe PDF



Page view(s)

checked on Feb 3, 2019


checked on Feb 3, 2019

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.