SET: a pupil detection method using sinusoidal approximation.

Amir-Homayoun Javadi,Vincent Walsh,Zahra Hakimi,Morteza Barati,Lili Tcheang

doi:10.3389/fneng.2015.00004

Abstract

Mobile eye-tracking in external environments remains challenging, despite recent advances in eye-tracking software and hardware engineering. Many current methods fail to deal with the vast range of outdoor lighting conditions and the speed at which these can change. This confines experiments to artificial environments where conditions must be tightly controlled. Additionally, the emergence of low-cost eye tracking devices calls for the development of analysis tools that enable non-technical researchers to process the output of their images. We have developed a fast and accurate method (known as “SET”) that is suitable even for natural environments with uncontrolled, dynamic and even extreme lighting conditions. We compared the performance of SET with that of two open-source alternatives by processing two collections of eye images: images of natural outdoor scenes with extreme lighting variations (“Natural”); and images of less challenging indoor scenes (“CASIA-Iris-Thousand”). We show that SET excelled in outdoor conditions and was faster, without significant loss of accuracy, indoors. SET offers a low cost eye-tracking solution, delivering high performance even in challenging outdoor environments. It is offered through an open-source MATLAB toolkit as well as a dynamic-link library (“DLL”), which can be imported into many programming languages including C# and Visual Basic in Windows OS (www.eyegoeyetracker.co.uk).

Highlights

We compared SET with the two previously mentioned open-source methods using two different sets of images: one extracted from videos taken in outdoor environments; the other from a dataset of images taken in controlled laboratory settings (“Chinese Academy of Sciences’ Institute of Automation (CASIA)-IrisThousand”)
A Two-Way ANOVA with method (SET/Starburst/Gaze-Tracker) and image collection (Natural/CASIA-Iris) as independent factors was conducted on processing time
We applied our method and two other commonly used open-source methods, Starburst (Li et al, 2005) and Gaze-Tracker (San Agustin et al, 2010) to two collections of eye images extracted from videos taken from outdoor (“natural outdoor scenes with extreme lighting variations (Natural)” image collection) and laboratory settings (“CASIA-Iris” image collection), and compared the performance of the methods in terms of detection rate, accuracy and speed

Summary

Introduction

In order to have ultimate freedom of movement and portability, a second group of eye-trackers were developed where the camera is mounted directly onto the head of the user (“mobile eye-trackers”), allowing the user to be free to walk around (e.g., www.ergoneers.com, Tobii and SMI). Most of these systems can be used in both indoor and outdoor applications. (2) Most pupil detection algorithms with high-precision suffer from trade-offs in speed or accuracy, e.g., Campadelli et al (2006) presented a fairly accurate method of localization of eyes in which the speed of processing is sacrificed for accuracy (4 s per frame) Most of these methods are based on an iterative algorithm. We compared SET with the two previously mentioned open-source methods using two different sets of images: one extracted from videos taken in outdoor environments; the other from a dataset of images taken in controlled laboratory settings (“CASIA-IrisThousand”)

Methods

Results

Findings

Discussion