Using Haar classifiers to detect driver fatigue and provide alerts

Drowsiness is a transition state between being awake and asleep and can have serious consequences when occurring in tasks that require sustained attention such as driving. During the state of drowsiness, reaction time is slower, vigilance is reduced, and information processing is less efficient, which may cause accidents. The proposed Driver Fatigue Detection System (called FDS) aims to monitor the alertness of drivers to prevent them from falling asleep at the wheel. The system monitors the driver’s face using Haar feature classifiers with an increased training set to detect changes in the face of the driver quickly. A correlation matching algorithm is used to accurately provide the target’s position and track the target’s eyes according to the intensity, shape, and size of the pupils. FDS uses an IR illuminator to produce the desired bright pupil effect when the driver is wearing sunglasses. The resulting system operates in real-time, and is more accurate and less intrusive to the driver than other systems currently available.