Real-Time Driver's Hypovigilance Detection using Facial Landmarks

Recently, driver hypovigilance (drowsiness and fatigue) becomes one of the principal causes of traffic crashes, it can prompt many deaths, wounds and many economic losses. Therefore, the use of a system that takes into account the driver's level of vigilance can play an important role in preventing accidents and saving human lives. In this work, we propose a non-intrusive driver hypovigilance detection system in real-time. This system makes it possible to detect drowsiness by the identification of MicroSleep corresponding to a sleepiness of more than 2 seconds through the analysis of eye-closure, and to identify fatigue by the analysis of the movement of the mouth to detect yawning. In case of drowsiness or fatigue, an alert is launched to make the driver vigilant and thus definitely avoid road accidents, decrease the percentage of murders and injuries caused by driver hypovigilance, then save many human lives. Experiments were conducted in real time to evaluate the proposed approach.