Table of Contents Author Guidelines Submit a Manuscript
Computational Intelligence and Neuroscience
Volume 2016, Article ID 2758103, 11 pages
http://dx.doi.org/10.1155/2016/2758103
Research Article

Classifying EEG Signals during Stereoscopic Visualization to Estimate Visual Comfort

1Université de Bordeaux, Potioc Project-Team, 351 Cours de la Libération CS 10004, 33405 Talence Cedex, France
2Inria, Inria Bordeaux Sud-Ouest, Potioc Project-Team, 200 Avenue de la Vieille Tour, 33405 Talence Cedex, France

Received 1 July 2015; Accepted 31 August 2015

Academic Editor: Stefan Haufe

Copyright © 2016 Jérémy Frey et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract

With stereoscopic displays a sensation of depth that is too strong could impede visual comfort and may result in fatigue or pain. We used Electroencephalography (EEG) to develop a novel brain-computer interface that monitors users’ states in order to reduce visual strain. We present the first system that discriminates comfortable conditions from uncomfortable ones during stereoscopic vision using EEG. In particular, we show that either changes in event-related potentials’ (ERPs) amplitudes or changes in EEG oscillations power following stereoscopic objects presentation can be used to estimate visual comfort. Our system reacts within 1 s to depth variations, achieving 63% accuracy on average (up to 76%) and 74% on average when 7 consecutive variations are measured (up to 93%). Performances are stable (62.5%) when a simplified signal processing is used to simulate online analyses or when the number of EEG channels is lessened. This study could lead to adaptive systems that automatically suit stereoscopic displays to users and viewing conditions. For example, it could be possible to match the stereoscopic effect with users’ state by modifying the overlap of left and right images according to the classifier output.