Fatigue Detection Based on Infrared Video Pupillography

Detection of drowsiness can be used in many fields to prevent accidents caused by human errors in machine and vehicle operations. The pupil which is innervated by the autonomic nervous system was found to change in parallel with the level of alertness. In this paper, we introduce a fatigue detection system based on infrared video pupillography. A subject's level of vigilance was distinguished by comparing the waveform of the varying pupile diameter of an alter person with a drowsy person's. The design of the system is illustrated in this paper, as well as the correct way to calculate the pupil size when the pupile boundary is occluded by eyelids or eyelashes. The results show that significant differences can be found in drowsy and alert groups: in the drowsy group, pupil diameter fluctuates with large amplitude at low frequencies; while in the alert group pupil size remains stable for a long time and oscillating with amplitudes rarely exceeding ±±5%(about 0.3mm).

[1]  H. Lüdtke,et al.  Mathematical procedures in data recording and processing of pupillary fatigue waves , 1998, Vision Research.

[2]  H.T. Nguyen,et al.  Early Driver Fatigue Detection from Electroencephalography Signals using Artificial Neural Networks , 2006, 2006 International Conference of the IEEE Engineering in Medicine and Biology Society.

[3]  M. Kusumi,et al.  The pupil as a possible premonitor of drowsiness , 2007, 2007 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society.

[4]  Prabir Bhattacharya,et al.  A driver fatigue recognition model using fusion of multiple features , 2005, 2005 IEEE International Conference on Systems, Man and Cybernetics.

[5]  H. Lüdtke,et al.  Daytime variations in central nervous system activation measured by a pupillographic sleepiness test , 2001, Journal of sleep research.