Driver drowsiness detection: a comparison between intrusive and non-intrusive signal acquisition methods

Driver drowsiness is a major cause of road accidents, many of which result in fatalities. A solution to this problem is the inclusion of a drowsiness detector in vehicles to alert the driver if sleepiness is detected. To detect drowsiness, physiologic, behavioral (visual) and vehicle-based methods can be used, however, only measures that can be acquired non-intrusively are viable in a real life application. This work uses data from a real-road experiment with sleep deprived drivers to compare the performance of driver drowsiness detection using intrusive acquisition methods, namely electrooculogram (EOG), with camera-based, non-intrusive, methods. A hybrid strategy, combining the described methods with electrocardiogram (ECG) measures, is also evaluated. Overall, the obtained results show that drowsiness detection performance is similar using non-intrusive camera-based measures or intrusive EOG measures. The detection performance increases when combining two methods (ECG + visual) or (ECG + EOG).

Introduction

Driving a car is a complex, multifaceted and potentially risky activity requiring both physiological and cognitive resources to maintain performance over time. A reduction in any of these resources can have dramatic consequences, resulting in crashes. Most people are aware of the dangers of driving while intoxicated, but many do not know that drowsiness also impairs judgment, performance and reaction times, just like alcohol and drugs do. The International Transport Forum at the OECD estimates that 20-30% of fatalities are due to driver sleepiness and fatigue [1], and in Europe, a survey across nineteen countries showed that the prevalence of falling asleep while driving in the previous 2 years is 17%, and amongst those who fell asleep, the prevalence of sleep-related crashes was 7% [2].