Title: Investigating classification methods to improve eye-tracking systems operation

Authors: Golnaz Baghdadi; Ali Motie Nasrabadi

Addresses: Biomedical Engineering Department, Shahed University, Tehran, Iran ' Biomedical Engineering Department, Shahed University, Across Emam Khomeini Holy Shrine, Persian Gulf Highway, Tehran, Iran

Abstract: There are different problems with eye-tracking systems such as users' head movements, multi-user compatibility and environment's light variation that reduce the accurately of these systems. The current research investigates the performance of using classifiers in the calibration stage of an infrared eye-tracking system on 10 subjects several times. The results show that classifiers can reduce the mentioned problems and improve the operation of the system to recognize the eye-gaze direction more accurately and furthermore give the system the ability to adjust with various interferential factors that will be interesting and important for a number of applications requiring human computer interfaces.

Keywords: human-computer interface; HCI; eye tracking technology; classification; infrared tracking; biomedical engineering; eye gaze direction.

DOI: 10.1504/IJBET.2012.049365

International Journal of Biomedical Engineering and Technology, 2012 Vol.10 No.2, pp.145 - 159

Published online: 12 Dec 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article