基於眼動儀及眼控自動校正功能之學習系統

A Learning System Based on Eye Tracker With Eye Gaze Auto Calibration

余舜華(Shun-Hua Yu)

指導教授 ： 王佳盈

中原大學/電機資訊學院/電機工程研究所/碩士(2021年)

https://doi.org/10.6840/cycu202100575

摘要

本論文主要利用眼動儀開發具有眼控自動校正功能的學習系統，希望能幫助早期療育兒童有效學習。作者接觸的早療機構中，有些兒童必須使用眼動儀進行眼部的訓練和課程的學習，而一般眼動儀在使用之前，必須先予以校正，但一些早療兒童受限於眼部肌肉的自我控制能力或者學習上的障礙，在注視螢幕焦點的時候，往往不容易維持穩定，造成校正操作的困難，以致眼動儀的使用效果不如預期。除此之外，市面上所開發的眼動應用通常是針對一般人所設計的，對於這些早療兒童來說節奏太快，而且辨識的圖案太小，也會進一步造成使用上的困難。
本論文設計的眼動訓練系統，採取使用時自動校正的方法。一開始只將螢幕畫面分割成左右兩邊，早療兒童可輕易辨識和使用。其後在持續使用過程中，系統會自動蒐集注視點的數據及建立辨識模型，同時循序漸進將螢幕畫面區分成更多等份。透過這樣的方式，系統可以達成自動校正的效果。我們希望這個系統未來能夠帶給早期療育兒童更多的幫助。

關鍵字

眼睛追蹤 ； 自動校正 ； 人機介面 ； 輔具應用

並列摘要

In this thesis, we use an eye tracker to design an auto-calibration eye-tracking learning system, the purpose is to help children in the early treatment institution we contacted. In the institution, some children must use an eye tracker for eye training and course learning. Generally, an eye tracker must be calibrated before normal use. However, due to limited eye muscle control or learning disabilities, these children often have difficulty maintaining a stable focus when looking at the screen, which can also cause calibration difficulties and unsatisfactory results. In addition, most eye-tracking applications are developed for ordinary people. For these early treatment children, the rate of change is too fast, and the recognized patterns are too small, which will further cause difficulties in use.
The proposed eye-tracking learning system adopts an automatic calibration method. At the beginning, the screen is only divided into left and right sides, so that children can easily recognize and use it. Afterwards, in the process of continuous use, the system will automatically collect the data and create a recognition model, and gradually divide the screen into more equal parts. In this way, the system can achieve the effect of automatic calibration. We hope that this system can bring more help to early treatment children who need eye trackers in the future.

並列關鍵字

Eye Tracking ； Automatic Calibration ； Calibration-Free ； Human Computer Interaction ； Assistive Technology

參考文獻

[1] 徐士理，「應用眼動儀於電動輪椅與機械手臂操控之實現」，碩士論文，中原大學電機資訊學院電機工程研究所，民國106年7月。

[2] 蔡東昇，「基於眼球注視偵測技術的自閉症孩童早期快篩系統」，碩士論文，中原大學電機資訊學院電機工程研究所，民國105年7月。

[3] Hackaday.io, “Patrick Joyce's Profile”. [Online]. Available: https://hackaday.io/Patrick.Joyce [Accessed: 26- Jun- 2021].

[4] Hackaday.io, “Eye Controlled Wheelchair!”. [Online]. Available: https://hackaday.io/project/5426-eye-controlled-wheelchair [Accessed: 26- Jun- 2021].

[5] Jiang, M., Francis, S., Srishyla, D., Conelea, C., Zhao, Q., and Jacob, S. Classifying Individuals with ASD Through Facial Emotion Recognition and Eye-Tracking. 41st Annual International Conference Of The IEEE Engineering In Medicine And Biology Society (EMBC), 2019, pp. 6063-6068.