Development of vehicle driver drowsiness detection system using electrooculogram (EOG)

Abstract

Driver drowsiness is one of the major causes of road accident. Various driver drowsiness detection systems have been designed to detect and warn the driver of impending drowsiness. Most available prototype and ongoing research have focused on video-based eye tracking system, which demands high computing power due to real time video processing. In our research, the use of electrooculogram (EOG) as an alternative to video-based systems in detecting eye activities caused by drowsiness is evaluated. The EOG, which is the electrical signal generated by eye movements, is acquired by a mobile biosignal acquisition module and are processed offline using personal computer. Digital signal differentiation and simple information fusion techniques are used to detect signs of drowsiness in the EOG signal. EOG signal is found to be a promising drowsiness detector, with detection rate of more than 80%. Based on the tested offline processing techniques, an online fatigue monitoring system prototype based on a Personal Digital Assistant (PDA) has been designed to detect driver dozing off through EOG signal.