Accurate and Robust Pupil Positioning Algorithm Using Adaboost Cascade Detector and Improved Starburst Model

DOI

10.2991/iccsae-15.2016.163How to use a DOI?

Keywords

Pupil positioning; eye tracking; Adaboost; starburst model.

Abstract

With the rapid development of human-computer interaction systems, eye tracking algorithm has become a very popular research field. As the most important part in eye tracking, pupil positioning algorithm is still facing many challenges, especially in processing non-ideal eye images. In this paper we propose an accurate and robust pupil positioning algorithm using Adaboost cascade detector and improved starburst model. Firstly, we use an average linear interpolating method to interpolate the specular reflections. Secondly, we combine HAAR features and Adaboost cascade detector to extract the region of interest (ROI) from the eye image (after specular reflections removal). Lastly, the improved starburst model is adopted to locate the pupil contour and the pupil center. In order to evaluate the performance of proposed method, we collect several eye moving videos using the specified camera device and the experimental results prove that the accuracy and robustness of proposed algorithm are better.

Copyright

© 2016, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Bing Liu
AU  - Zhiyi Qu
AU  - Yueqing Ren
AU  - Ruicheng Wang
PY  - 2016/02
DA  - 2016/02
TI  - Accurate and Robust Pupil Positioning Algorithm Using Adaboost Cascade Detector and Improved Starburst Model
BT  - Proceedings of the 2015 5th International Conference on Computer Sciences and Automation Engineering
PB  - Atlantis Press
SP  - 877
EP  - 883
SN  - 2352-538X
UR  - https://doi.org/10.2991/iccsae-15.2016.163
DO  - 10.2991/iccsae-15.2016.163
ID  - Liu2016/02
ER  -