Drowsiness Detection and Analysis

Author: Yi Shen

Shen, Yi, 2017 Drowsiness Detection and Analysis, Flinders University, School of Computer Science, Engineering and Mathematics

This electronic version is made publicly available by Flinders University in accordance with its open access policy for student theses. Copyright in this thesis remains with the author. This thesis may incorporate third party material which has been used by the author pursuant to Fair Dealing exceptions. If you are the owner of any included third party copyright material and/or you believe that any material has been made available without permission of the copyright owner please contact copyright@flinders.edu.au with the details.


Driver fatigue is one of major reasons which causes vehicle accidents. The drivers’ drowsiness detection can be described by ocular measures, such as eye blink duration and blink frequency. The aim of this project is to use ocular measures to detect and analyse drowsiness in real-time. This project consists of four parts which are eye detection, eye tracking, eye blink detection and drowsiness detection respectively. Eye detection used the Viola-Jones algorithm which is based on cascade Haar classifiers to identify region of interest from images. Eye tracking combined the Camshift algorithm with KLT algorithm to track eye pair of participants in the real-time detection. Eye blink detection used the correlation coefficient algorithm to calculate correlation scores between the open eyes template image and the region of interest of every frame. The analysis of eye blink detection measured the blink duration, blink frequency in drowsy, alert, normal state. Also, it compared the differences of blink duration and frequencies in different drowsiness states. Furthermore, it analysed factors which affect the accuracy of drowsiness detection.

Keywords: Drowsiness detection, Classifier, Camshift, KLT, Blink frequency, Blink duration.
Subject: Engineering thesis

Thesis type: Masters
Completed: 2017
School: School of Computer Science, Engineering and Mathematics
Supervisor: Sherry Randhawa