Objective Evaluation of Driver Fatigue by Using Spontaneous Pupillary Fluctuation

Methods for objective and reliable evaluation of driver fatigue are important means to suppress fatigued driving and prevent traffic accident. In this paper, we presented an effective driver fatigue detection method which based on spontaneous pupillary fluctuation behavior. The first step consists of acquisition of a 90s long infrared video of pupillogram by using infrared-sensitive CCD camera. Second, we employed edge detection algorithm based on curvature characteristic of pupil boundary to extract a set of points of visible pupil boundary, particularly, the eyelids or eyelashes are occluding the pupil. Then we adopted these points to fit a circle by using least squares error criterion, the diameter of this fitted circle is the approximated diameter of the pupil in current frame of video. Finally, the values of pupil diameter variability (PDV) in 90s long video was calculated. In a preliminary study, the PDV of two groups were measured. One group rated themselves as alert (12 men), the other group as sleepy (13 men).The results show that significant differences between alert groups and drowsy ones, the values are 0.026 + 0.005 and 0.141+0.042, respectively. Because spontaneous pupillary fluctuation is innervated by autonomic nervous system whose activity is known to change in parallel with drowsiness and cannot be influenced by subjective motive of people. When combined with the results, we could draw a conclusion that PDV was a potential indicator that can be used for objective evaluation of driver fatigue.