This paper adopts CCD video fixed in a cab monitoring driver to recognize whether he/she is tired. The drivers' alertness level is detected by monitor both their eyes and mouths states. This paper puts forward a new method to locate the driver's face feature points by using pigment distributing. First, according to the character that lip is red, detecting and searching the red region in order to separate the lip from other parts, then combine the edge extract and red pigment method to get the shape of driver's lip. Then use the approaching method to locating the lip's feature points. The eye's position is fixed according to human face figure and its feature points are obtained by using the same method with lip. Put all the feature points into BP networks, adopts area matching arithmetic to recognize driver's alertness level, fatigue monitor system can recognize that the driver is in decreased alertness state when he/she yawns more times than normal, then the system gives off alarm signal, and awakes the driver; me system can also recognize that the driver is in drowsy state when driver's eyes blink frequency is lower than standard value, then it gives sharp alarm signal to wake the driver up. It has the characteristic of real time, accuracy. The recognized accuracy rate reaches to 94.3%. This driver fatigue monitoring system has significant effect to reduce traffic accident.
Driver's Alertness Level Identify Method Based on Computer Vision
First International Conference on Transportation Engineering ; 2007 ; Southwest Jiaotong University, Chengdu, China
2007-07-09
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
Driver's Alertness Level Identify Method Based on Computer Vision
British Library Conference Proceedings | 2007
|Estimate of Driver's Alertness Level Using Fuzzy Method
SAE Technical Papers | 1994
|Correlation between vehicle's meandering and driver's alertness level
Tema Archiv | 1995
|Active Safety Improvement through Effective Use of Driver's Alertness level Information
British Library Conference Proceedings | 2000
|Analysing Driver's Attention Level using Computer Vision
IEEE | 2008
|