This content is not included in your SAE MOBILUS subscription, or you are not logged in.
The Estimation of Driver's Arousal States (2)~Based on Blink Video Sequence
Published May 21, 2008 by Society of Automotive Engineers of Japan in Japan
Event: JSAE Spring Meeting
We propose a real-time method of the estimation of driver's arousal states based on blink waveform categorization and histogram variation. The driver's arousal level while driving is not monotonous falling from high arousal level to low. We assumed an arousal state transition model which has taken into account the fact that a driver usually held out against sleepiness while driving. We propose the estimation method of driver's arousal states based on blink video sequence, which can be applied for unspecified drivers. The blink pattern categories were classified from blink video sequences using HMM (Hidden Markov Model), then the driver's arousal states were finally estimated using HMM by histogram variation of those typical blink categories.