| 摘要 |
PROBLEM TO BE SOLVED: To determine a drowsy state with high accuracy by extracting a feature quantity of blinking eyes using a threshold suited to a type of the feature quantity of the blinking eyes. SOLUTION: An image pick-up device 12 photographs a region including the eyes of a driver and an eye opening degree detection section 24 detects an eye opening degree. A feature quantity extraction section 30 extracts a plurality of types of blinking feature quantities based on time-series data of the detected eye opening degree using a standard threshold of closed eyes, large thresholds of the closed eyes and small thresholds of closed eyes stored in a threshold storage section 28. A threshold determination section 32 determines the thresholds of the plurality of types of feature quantities of blinking eyes, and a drowsiness determination section 34 determines the drowsy state of the driver based on the result of the threshold determination of the plurality of types of feature quantities of the blinking eyes. COPYRIGHT: (C)2010,JPO&INPIT
|
