臺灣博碩士論文加值系統

本篇論文使用眼動儀建立一個人機介面，透過眼動儀取代滑鼠的功能，讓無法自由活動、溝通之病患可以利用該系統傳遞身體資訊給醫療照護人員，並且可透過該系統與他人進行簡單的溝通。近年來眼球追蹤技術快速的成長，其計算能力比以前更加快速且準確，可提供人機介面更加準確的控制。但大多數的眼動儀的數據都使用於心理分析，像是偵測使用者最常注視的範圍。透過建立身體資訊傳遞功能，將身體特徵圖分別建立於系統畫面，病患透過眼球移動的方式控制滑鼠選擇身體特徵部位，進而將身體資訊快速且正確傳達給醫護人員。搭配常用字語功能，可以與家人、朋友間進行簡單之溝通，並且可依病患需求客製化系統。另外透過設定閥值去判別是否觸發滑鼠Click事件。身體資訊傳遞功能可有效並且快速的與醫護人員傳遞身體異常部位，讓醫護人員在病患處理狀況之時除了透過一般醫療器材所得知之資料訊息外，還可快速並正確地獲取病患本身所傳遞之異常部位，可提升判斷其異常狀況之精準度並快速應變排除。而常用字語功能則提供病患透過選取常用字語與家人、朋友們進行簡單之互動，可提升病患之日常生活品質，並且該功能具有高度的靈活性，可根據病患之要求進行客製化的功能，以提供病患更有效率的使用該功能。而且透過設定閥值來控制滑鼠Click事件可大幅增加準確率，且可有效降低系統誤判率，進而大幅提升系統正確使用效率。

This study is the use of eye tracker to develop a human-computer interaction system. Let disabled patient can transfer their body information to nursing aide and communicate with people by using the eye tracker. In recent years, the eye tracking technology has been grown up and the calibration ability of eye tracking is faster and more precise than before. The eye tracking can provide a precise control at human-computer interaction. Most eye trackers are developed for psychology analysis by collecting eye gaze data, e.g., detecting the area where a user looks most. We develop a system interface called “Body Information Transmission Function” that placed a human body image, and let patient gaze the area of body feature to transmit the information to nursing aide. With the “Common Words Function”, patient can do simple communicate with his family or friends. This function also can customize the interface according to the patient’s request. Besides, we set the threshold to trigger the mouse click event. The body information transmission function can quickly let patient transmits his abnormal body parts to nursing aide. Let nursing aide can get the information through normal medical equipment, but also can quickly and accurately obtain the abnormal body parts from patients themselves when they deal with patients problem. It can improve the accuracy and response time of the abnormal condition of the judgment that. And, let nursing aide can deal with the problem in a quickly time. The common words function provides the patient doing simple communication with his family and friends by selecting the common words. It can improve the quality of patient’s daily life, and the high flexibility of this function provides customizing system according to patient’s request. Let patients can use this function more effectively.
Through setting the threshold of selecting time, the input method of this system can improve the accuracy of input and decrease the probability of error input. It im-proves the effective of system usability.

誌謝 i
摘要 ii
Abstract iii
Chapter 1 INTRODUCTION 1
1.1 Background 1
1.2 Motivation 2
Chapter 2 LITERATURE REVIEW 3
2.1 Human-Computer Interaction（HCI） 3
2.2 Eye-based Human-Computer Interaction 3
2.2.1 Limbus Tracking Method 4
2.2.2 EOG Method 4
2.2.3 Pupil Detection Method 5
2.3 Eye Tracker 6
Chapter 3 Research Method 8
3.1 Eye Gaze Tracker System 8
3.2 Eye Gaze Data Processing 10
3.3 System Design 11
3.3.1 Body Information Transmission Function 11
3.3.2 Common Words Function 12
3.4 Input Method 13
Chapter 4 System Performance 14
4.1 Experimental Data 15
Chapter 5 Conclusions and Future Work 27
References 29

[1]Abe, K., Ohi, S., & Ohyama, M. (2011). Eye-gaze Detection by Image Analysis under Natural Light. In J. Jacko (Ed.), Human-Computer Interaction. Interaction Techniques and Environments (Vol. 6762, pp. 176-184): Springer Berlin Heidelberg.
[2]Arai, K., & Mardiyanto, R. (2011a, 11-13 April 2011). Eye-based HCI with Full Specification of Mouse and Keyboard Using Pupil Knowledge in the Gaze Estimation. Paper presented at the Eighth International Conference on Information Technology: New Generations (ITNG).
[3]Arai, K., & Mardiyanto, R. (2011b, 11-13 April 2011). Eye Based HCI with Moving Keyboard for Reducing Fatigue Effects. Paper presented at the Eighth International Conference on Information Technology: New Generations (ITNG).
[4]Chuan, N. K., & Sivaji, A. (2012, 3-4 Dec. 2012). Combining eye gaze and hand tracking for pointer control in HCI: Developing a more robust and accurate interaction system for pointer positioning and clicking. Paper presented at IEEE Colloquium on the Humanities, Science and Engineering (CHUSER).
[5]Dix, A. (2004). Human-computer Interaction: Pearson/Prentice-Hall.
[6]Jacob, R. J. K. (1996). Human-computer interaction: input devices. ACM Comput. Surv., 28(1), 177-179. doi: 10.1145/234313.234387
[7]Krejcar, O. (2011, 5-7 Oct. 2011). Handicapped people virtual keyboard controlled by head motion detection. Paper presented at International Congress on the Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT).
[8]Parmar, K., Mehta, B., & Sawant, R. (2012, 19-20 Oct. 2012). Facial-feature based Human-Computer Interface for disabled people. Paper presented at International Conference on the Communication, Information & Computing Technology (ICCICT).
[9]Sinha, G., Shahi, R., & Shankar, M. (2010, 19-21 Nov. 2010). Human Computer Interaction. Paper presented at the International Conference on Emerging Trends in Engineering and Technology (ICETET).
[10]Usakli, A. B., & Gurkan, S. (2010). Design of a Novel Efficient Human-Computer Interface: An Electrooculagram Based Virtual Keyboard. IEEE Transactions on, Instrumentation and Measurement, 59(8), 2099-2108.