Detection of consciousness degradation and concentration of a driver for friendly information service

In order to provide a friendly information service to drivers, the state of the driver's consciousness must be known. The authors have developed a basic method for detecting the consciousness degradation and concentration of a driver by using image processing. The consciousness degradation of a driver is inferred from the change in blink duration determined from facial images. The blink recognition rate obtained in animated image processing was over 95% and the consciousness degradation was estimated from the changes over tune of the average duration of eye closure. The authors have prospects for detecting the concentration level from the two-dimensional distribution of gaze direction.