Eye tracking system

Abstract

Robust, non-intrusive human eye tracking problem has been a fundamental and challenging problem for computer vision area. The aim of this project is to design and develop a real-time, robust, non-intrusive eye tracking system with human eye movement indication property using the movements of eye pupil. The eye-tracking algorithm, called EyeTracer, is implemented using the Continuously Adaptive Mean-Shift (CAMSHIFT) algorithm developed by Bradski, G. Human eye image captured by the webcam is detected using the CAMSHIFT algorithm which draws an ellipse around the circumference of the pupil and tracks its movement. EyeTracer is able to continuously track and determine the coordinates of the eye; texture and pictorial data can be recorded and saved for future analysis use. The results show that EyeTracer can deliver more than 85% of accuracy in tracking and determining eye’s movement and fixation.