追瞳系統之研製及其應用

The Development of an Eye-Tracking System and Its Applications

邱國鈞(Kuo-Chun Chiu)

指導教授 ： 蘇木春

國立中央大學/資訊電機學院/資訊工程學系/碩士(2006年)

摘要

追瞳系統是一種可以偵測人類眼睛注視方向的系統。藉由偵測人類眼睛視線，可以明確顯示使用者感興趣的地方，因此追瞳系統也被廣泛應用於人機介面之使用，可以判斷分析眼睛的疾病，同時，也可以用在人類認知學的研究。因此，眼睛視線追蹤成為近年來，成為近幾年來最具吸引力的研究之一。追蹤眼睛移動方向有很多不同的方法，例如利用光線在眼睛上造成反射的紅外線眼動圖法，或是利用生理訊號（EOG）的眼電圖法…等。然而每一種方法都有其優缺點，例如有些已經商品化的系統，他們的價錢通常都非常的昂貴；另一方面，有些系統需要貼上額外的電極片。此外，還有的系統是要求使用者的頭部，在完全不能移動的狀態下才能夠使用；除此之外，複雜的校正流程常常需要不斷的執行，在使用上造成極大的不方便。
在本篇論文中，主要是提供一套低價位的追曈系統，透過系統的使用，可以讓身障者利用眼睛的視線來操控電腦。系統中使用兩顆低價位之視訊攝影機，將第一顆視訊攝影機架設在耳機上，並且在攝影機旁加裝兩顆紅外線LED，讓使用者戴在頭上使用，而第二顆視訊攝影機則置於電腦螢幕的下方。利用第一台視訊攝影機擷取眼睛畫面，並將擷取之影像，計算使用者眼睛注視的方向。研發的追瞳演算法，能有效的計算影像中瞳孔中心位置，並且追蹤瞳孔的移動。接著利用簡單且有效的對映方法，計算出眼睛注視的方向。第二台攝影機則是用來補償因使用者頭部移動而產生的偏移量。部分身障者如漸凍人（肌萎縮性側索硬化症），可透過這套追曈系統以及搭配研發的應用程式，可以呼叫旁人的協助，中英文文字輸入，瀏覽網際網路或是控制家中電器等功能。最後，經由實驗結果可以得知，我們的追瞳系統可以快速且精確地執行追蹤瞳孔的工作。

關鍵字

人機介面 ； 漸凍人 ； 頭戴式攝影機 ； 追瞳系統 ； 輔具

並列摘要

Eye gaze tracking systems are systems that can estimate the gaze direction of a human user. Since eye gaze direction can express the interest of a user the applications of an eye gaze system are widely varied from human computer interaction and eye disease diagnosis to human cognition study. Therefore, eye gaze tracking has been a very attractive research topic in recent years. There are several different ways to track the directions of eye movements by refection of light and electrooculographic potential (EOG), etc. Each approach has its own advantages and disadvantages. For example, some systems are already commercially available; however, the costs of these systems usually are very expensive. On the other hand, some systems require attaching electrodes. Furthermore, some systems require the user to keep his or her head almost completely still; otherwise, a complex calibration process needs to be repeated again and again.
In this thesis, a low cost eye gaze tracking system is proposed. With the eye gaze tracking system, people with severe disabilities can access computers via eye gaze. The system consists of two low-cost Web cameras. While the first Web camera surrounded by two infrared LEDs is attached to a headset which is worn on the user’s head the second Web camera is under the computer monitor. The images acquired by the first Web camera are used to compute eye gaze directions. An efficient gaze tracking algorithm has been developed to locate the center of pupil in images and then track the movements of the pupil. Then a simple but effective eye gaze mapping algorithm is used to compute eye gaze directions. The images acquired by the second Web camera are used to compensate head movements. Via the eye gaze tracking system and application software, people with severe disabilities, such as severe cerebral palsy or amyotrophic lateral sclerosis (ALS), can ask for helps, type texts, browse Internets, and control home appliances etc. Experimental results demonstrate the performance of the proposed system.

並列關鍵字

eye gaze ； eye-tracking system ； head-mount camera

參考文獻

[39] 李佳霖，「頭配式顯示器之人體生理參數量測分析」，私立中原大學，醫學工程研究所碩士論文，民國九十年六月。

[8] R. L. Hsu, M. A. Mottaleb, and A. K. Jain, “Face detection in color images,” IEEE Transactions on Pattern Analysis and Machine Intell., vol. 24, pp. 696-706, 2002.

[1] J. L. Andreassi, Psychophysiology: Human Behavior and Physiological Response, Third Edition. Hillsdale, NJ: Lawrence Erlbaum, 1995.

[2] M. Betke, J. Gips, and P. Fleming, “The camera mouse: visual tracking of body feature to provide computer access for people with severe disabilities,” IEEE Transaction on Neural Systems and Rehabilitation Engineering, vol. 10, no. 1, pp. 1-10, 2002

[3] R. H. S. Carpenter, Movements of the eyes, Aug. 1988