This content is not included in your SAE MOBILUS subscription, or you are not logged in.
The Assessment of Driver's Arousal States From Video Sequences Based on the Classification of Eye-Blink Patterns
Published May 23, 2007 by Society of Automotive Engineers of Japan in Japan
Event: JSAE Spring Conference
The assessment of driver's arousal levels is one of the most important issues for the development of an adaptive driving support system. This paper proposes a novel nonintrusive approach to assess driver's arousal levels by analyzing eye-blink patterns from the recorded in-vehicle video data. Drowsy drivers involuntarily produce unique blink pattern characteristics. The proposed method uses Hidden Markov Models (HMMs) to classify blink patterns, and the drowsiness levels are determined from temporal transitions of these blink patterns. Further analysis also reveals a strong correlation between the eye-blink patterns derived from this approach and those derived from the recorded EOG (electro-occulography) waveforms.