本篇論文中，主要的目的是追蹤凝視者之雙眼位置並且加以判斷眼睛的狀態是否睜開或是閉上。而在本文之前，已經有許多學者針對眼睛的追蹤做了許許多多不同的研究，大部份的學者都是專注於如何將眼睛區域裡的各種眼睛之特性粹取出來 (例如虹膜與上眼瞼和下眼瞼的邊緣、瞳孔與眼白的顏色、瞳孔對紅外線的反射等 )，再作為眼睛位置與狀態之判斷準則；如此截取眼睛部份來做追蹤依據，在睜開眼睛時，可以追蹤的很好，但是當閉上眼睛或是半閉合時，由於參數不明確，容易把眉毛當成閉眼來判斷。

本篇研究採用追蹤眉毛（brow tracking）的方式，就所謂〝山不轉，路轉〞，由於眉毛特性算是剛體，只有在憤怒與悲傷時有所短暫的變化，比起眼睛更為固定，所以使用眉毛來當眼睛位置的主要依據；而方法則是利用粒子濾波器（particle filter）來追蹤兩眉毛的中心，在使用左右眉毛的樣板比對左右兩眉毛的精確位置，再對眼睛之睜眼閉眼做狀態評估，最後使用眼睛樣板找出眼睛虹膜和眼眶位置，如此能達到精確又好的效果。而在追蹤前的初步眼睛定位，則使用膚色與人臉邊緣先將人臉偵測出來再進行對比度分析，即可快速的取得眼睛的區域，最後使用可變型樣板（deformable template）將眼睛定位出。

In this thesis, our main purpose is to track the location of human eyes and to detect the state of eyes, which can be open or closed. Many researchers had investigated the problem of eyes tracking before this thesis. Most of them are focused on how to retrieve the properties of eyes from eyes region (e.g. the edge of the eyelid and iris, the colors of the cornea and pupil, reflected infrared light for bright pupil, etc.). Those properties are then used to detect the location and the state of the eyes. It works well when the eyes are open. However, when the eyes are closed or blinking, eyebrows may be mistaken as closed eyes, thus resulting in errors.

We propose a method that tracks eyes based on brows in this research. It simplifies the tracking step. The shape of a brow rarely changes except when a person feels anger or deep sorrow. It is more rigid than the eye. We try to take advantage of this fact to locate human eyes by locating the brows above them first. This method is to track between the eyebrows by a particle filter, and to use the template of right and left brow to match the accurate location of right and left brow. Then, we detect the eye state by their histogram. Finally, the exact position of the iris and eyelid can be found by deformable template. Before we start tracking mode, the eye acquisition must be detected by analyzing contrast of human face, which can be derived from skin-tone and edge of human face. Finally, eyes acquisition will be found by deformable template. Experiment show that our method can effectively track eyes and detect eye state.

[1] R.C. Gonzalez, R.E. Woods, Digital Image Processing, ed., Prentice Hall, 1987.

[2] S. Birchfield, “Elliptical Head Tracking Using Intensity Gradient and Color Histograms,” In Proc. IEEE Conf. CVPR, PP. 232–237, Jun. 1998.

[3] H. Rowley, S. Baluja, and T. Kanade, “Neural Network-Based Face Detection,” IEEE Trans. PAMI, vol. 20, no. 1, pp. 23-38, 1998.

[4] C.-C. Han, H.-Y. Mark Liao, G.-J. Yu, and L.-H. Chen, “Fast face detection via morphology-based pre-processing,” Pattern Recognition,vol. 33,
pp. 1701-1712, 2000.

[5] S. Kawato and J. Ohya.“between-eyes,”Proc. IEEE 4th Int. Conf. on Automatic Face and Gesture Recognition, pp. 40-45, Mar. 2000.

[6] Qiang Ji and Zhiwei Zhu, “Eye and gaze tracking for interactive graphic display,” Machine Vision and Application, Vol. 15, pp. 139-148, July 2004.

[7] Young-Sheng Chen, Chan-Hung Su, Jiun-Hung Chen, Chu-Song Chen, Yi-Ping Hung, and Chiou-Shann Fuh, “Video-based Eye Tracking for Autostereoscopic Displays,” Optical Engineering, Vol.40, no.12, pp. 2726-2734, Dec. 2001.

[8] H.Kasper, B.J.M.Hess, “Magnetic Search Coil System for Detection of Three-Dimensional Angular Movements” IEEE Trans. Vol.38 no.5 pp.466-475 , May 1991.

[9] Qiuping Ding, Kaiyu Tong, and Guang Li, “Development of an EOG (Electro-Oculography) Based Human-Computer Interface” IEEE Engineering in Medicine and Biology 27th Annual Conference Shanghai, China, 2005.

[10] Zhiwei Zhu, Qiang Ji,“Robust Real-time Eye Detection and Tracking Under Variable Lighting Conditions and Various Face Orientations,”Computer Vision and Image Understanding, , vol. 38, no. 1, PP. 124-154, 2005.

[11] M. Isard and A. Blake ,“Condensation - Conditional Density Propagation for Visual Tracking,”International Journal of Computer Vision, vol. 29, no. 1, pp. 5-28, 1998.

[12] J. Carpenter, P. Clifford, and P. Fearnhead, ”Improved particle filter for nonlinear problems,” IEE Proc. F, vol. 146, pp. 2-7, 1999.

[13] Rein-Lien Hsu, Mohamed Abdel-Mottaleb, and Anil K. Jain, “Face Detection in Color Images,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 24, no. 5, pp. 696-706, May 2002.

[14] D. Chai and K.N. Ngan, “Face segmentation using skin color map in videophone applications,” IEEE Transactions on Circuits and systems for Video Technology, U.S.A., vol. 9, no. 4, pp. 551-564, June 1999.

[15] A.L.Yuille, D.S.Cohen, and P.W.Hallinan, “Feature Extraction from Faces Using Deformable Template,” IEEE Proc., pp.104-109, June 1989.

[16] Y.Wu, H.Liu, and H.Zha. “A new method of detecting human eyelids based on deformable templates,” In Systems, Man and Cybernetics, vol. 1, pp. 604–609, Oct. 2004.

[17] Feng Liu , Qingshan Liu , Hanqing Lu, “Robust Color-Based Tracking,” Third International Conference on Image and Graphics (ICIG'04), pp. 132-135, Dec. 2004.

[18] 張凱傑，“眼控與頭控之人機介面系統研發與整合”，逢甲大學自動控制工程研究所碩士論文，民國九十年。

[19] 賴梓雍，“基於黑白立體視覺之眼睛偵測”，國立台灣科技大學，電子工程研究所，碩士學位文，2006。