基于人眼运动特征的眼动操作系统若干关键技术研究

发布时间：2018-03-21 07:03

本文选题：人机交互　切入点：眼动操作系统　出处：《天津大学》2012年硕士论文　论文类型：学位论文

【摘要】：随着科技的发展,眼动操作系统作为高级人机交互的重要应用之一,逐渐成为当前国内外相关科研领域的热点课题,已经被应用到医学、心理学、广告学等领域,尤其在军事、航空航天等方面具有重要的意义。目前,眼动操作系统仍处于研究的初级阶段,因此对于其关键技术及控制信息指标的研究具有重要的实践意义和理论价值。本文概括叙述了眼动操作系统技术的发展现状,总结了当前眼动操作系统在发展中所遇到的问题及其发展方向,分析了眼动操作系统的整体概念、人眼结构及眼睛的基本参数指标等基础知识进行,并根据眼睛运动基本形式的数据范围对眼动操作系统的结构及性能要求进行了研究与分析。在眼动操作系统中,将人眼运动的基本形式作为眼动操作新系统信息输入部分的输入信息,包括注视、眼跳、平滑尾随跟踪三种,其中注视作为眼睛运动中捕捉外界信息的最重要途径,本文对其进行了详细深入的分析。眼动跟踪技术是眼动操作系统中信息输入部分的关键技术,也是整个眼动操作系统的核心技术之一,本文对其进行了深入探索研究,在前人研究的基础上对眼动跟踪技术进行了重新分类,同时,对该技术的2D及3D两种方式作出比对,分析了两者的不同应用场合。本文根据眼动操作系统的结构,详尽的分析了控制输入部分的两个关键控制信息,并结合眼动仪的先进手段进行实验验证。第四章中深入研究了现阶段常用眨眼检测方法,并对眨眼控制信息进行了分析和实验验证,设计实验检测眼睛疲劳对眨眼控制的影响,最终得到对眨眼控制精度影响最小的注视时间阈值;第五章对于控制输入部分的另一个关键控制信息——注视时间进行了详细深入的研究,首先通过分析发现在人眼的注视过程中,眼睛会不自主的进行无规则的抖动,而这种抖动的范围具有一定的规律且其大小不可忽视,因此本文对注视抖动进行了的实验设计与结果分析,实验结果表明注视抖动的范围在1°以内,设计眼动操作系统时应予以注意,最后通过实验验证了眼睛疲劳是否对注视位置精度有所影响。
[Abstract]:With the development of science and technology, eye movement operating system, as one of the important applications of advanced human-computer interaction, has gradually become a hot topic in the field of scientific research at home and abroad, and has been applied to the fields of medicine, psychology, advertising and so on, especially in the military. At present, the eye movement operating system is still in the primary stage of research, so it has important practical significance and theoretical value for the research of its key technology and control information index. This paper briefly describes the development of eye movement operating system technology, summarizes the problems encountered in the development of eye movement operating system and its development direction, and analyzes the overall concept of eye movement operating system. The basic knowledge of the human eye structure and the basic parameter index of the eye is carried out, and the structure and performance requirements of the eye movement operating system are studied and analyzed according to the data range of the basic form of eye movement. The basic form of human eye movement is regarded as the input information of the information input part of the new system of eye movement operation, which includes three kinds: gaze, eye jump and smooth trailing tracking, among which gaze is the most important way to capture the external information in eye movement. This article has carried on the detailed thorough analysis to it. Eye movement tracking technology is the key technology in the information input part of eye movement operating system, and it is also one of the core technologies of the whole eye movement operating system. On the basis of previous researches, the eye movement tracking technology is reclassified. At the same time, the 2D and 3D methods of this technique are compared, and the different applications of the two methods are analyzed. According to the structure of eye movement operating system, two key control information of control input part are analyzed in detail in this paper, and the experimental verification is carried out with the advanced means of eye movement instrument. In chapter 4th, the commonly used blink detection methods are deeply studied. The blinking control information is analyzed and verified by experiments. The effect of eye fatigue on blink control is tested experimentally. Finally, the fixation time threshold, which has the least effect on blink control accuracy, is obtained. In Chapter 5th, another key control information in the control input, fixation time, is studied in detail. First of all, it is found that in the process of human eye gaze, the eye unconsciously jitter. However, the range of this jitter has certain regularity and its magnitude can not be ignored. Therefore, the experimental design and analysis of the results of this paper show that the range of fixation jitter is less than 1 掳. Attention should be paid to the design of eye movement operating system. Finally, the effect of eye fatigue on the accuracy of gaze position is verified by experiments.
【学位授予单位】：天津大学
【学位级别】：硕士
【学位授予年份】：2012
【分类号】：TP11;R318.0

【参考文献】