Abstract

Today, number of accidents happen during drowsy driving on roads and are increasing day by day. It is a known fact that many accidents occur due to driver’s fatigue and sometimes due to inattention factor. This research mainly engages on maximizing the effort in identifying the drowsiness state of driver in real driving conditions. The goal of driver drowsiness detection systems is an attempt to contribute in reducing these road accidents. The secondary data collected focuses on past research on drowsiness detection systems and various methods have been used earlier for detection of drowsiness or inattention while driving. However, in this paper, a real time vision-based method is proposed to monitor driver fatigue. This research approach adopts the Viola-Jones classifier to detect the driver’s facial features. Firstly, the face is located by a Haar like feature based object detection algorithm. The face area is detected using the functions in the OpenCV library with C#.net. Secondly, eye is detected. Also the eye areas are detected by using the functions in the OpenCV library and tracking by using a template matching method. Then, the open/close state of eyes is determined, and then fatigue is determined based on the series state of eyes. The correlation coefficient template matching method is applied to derive the state of each feature on a frame by frame basis. Vision- based driver fatigue detection method is a natural, non-intrusive and convenient technique to monitor driver’s vigilance.

Authors and Affiliations