Quantitative Measurement of Eyestrain on 3D Stereoscopic Display Considering the Eye Foveation Model and Edge Information

被引：20

作者：

Heo, Hwan ^{[1
]}

Lee, Won Oh ^{[1
]}

Shin, Kwang Yong ^{[1
]}

Park, Kang Ryoung ^{[1
]}

机构：

[1] Dongguk Univ, Div Elect & Elect Engn, Seoul 100715, South Korea

来源：

SENSORS | 2014年 / 14卷 / 05期

基金：

新加坡国家研究基金会;

关键词：

eyestrain; gaze estimation; edge strength; 3D stereoscopic display; VISUAL FATIGUE; GAZE TRACKING; SYSTEM; ACCOMMODATION; RESPONSES; IMAGES;

D O I：

10.3390/s140508577

中图分类号：

O65 [分析化学];

学科分类号：

070302 ; 081704 ;

摘要：

We propose a new method for measuring the degree of eyestrain on 3D stereoscopic displays using a glasses-type of eye tracking device. Our study is novel in the following four ways: first, the circular area where a user's gaze position exists is defined based on the calculated gaze position and gaze estimation error. Within this circular area, the position where edge strength is maximized can be detected, and we determine this position as the gaze position that has a higher probability of being the correct one. Based on this gaze point, the eye foveation model is defined. Second, we quantitatively evaluate the correlation between the degree of eyestrain and the causal factors of visual fatigue, such as the degree of change of stereoscopic disparity (CSD), stereoscopic disparity (SD), frame cancellation effect (FCE), and edge component (EC) of the 3D stereoscopic display using the eye foveation model. Third, by comparing the eyestrain in conventional 3D video and experimental 3D sample video, we analyze the characteristics of eyestrain according to various factors and types of 3D video. Fourth, by comparing the eyestrain with or without the compensation of eye saccades movement in 3D video, we analyze the characteristics of eyestrain according to the types of eye movements in 3D video. Experimental results show that the degree of CSD causes more eyestrain than other factors.

引用

页码：8577 / 8604

页数：28

共 34 条

[1]

[Anonymous], 2006, Digital Image Processing

[2]

Ardouin J., 2011, Proceedings 2011 IEEE Symposium on 3D User Interfaces (3DUI 2011), P95, DOI 10.1109/3DUI.2011.5759223

[3] High frequency edges (but not contrast) predict where we fixate: A Bayesian system identification analysis [J].