Driver fatigue monitoring system using Support Vector Machines

Abstract

Driver fatigue is one of the leading causes of traffic accidents. This paper presents a real-time non-intrusive fatigue monitoring system which exploits the driver's facial expression to detect and alert fatigued drivers. The presented approach adopts the Viola-Jones classifier to detect the driver's facial features. The correlation coefficient template matching method is then applied to derive the state of each feature on a frame by frame basis. A Support Vector Machine (SVM) is finally integrated within the system to classify the facial appearance as either fatigued or otherwise. Using this simple and cheap implementation, the overall system achieved an accuracy of 95.2%, outperforming other developed systems employing expensive hardware to reach the same objective.