以類神經網路為基礎之眼動追蹤系統｜國立中央大學博碩士論文系統

簡易檢索 / 詳目顯示

回結果列表

研究生：	余達明 U TAT MENG
論文名稱：	以類神經網路為基礎之眼動追蹤系統 A Neural-networks-based Eye-Tracking System
指導教授：	蘇木春 Mu-Chun Su
口試委員:
學位類別：	碩士 Master
系所名稱：	資訊電機學院 - 資訊工程學系 Department of Computer Science & Information Engineering
論文出版年：	2017
畢業學年度：	105
語文別：	中文
論文頁數：	106
中文關鍵詞：	眼動追蹤、類神經網路、Inner Corner-Pupil Center Vector
外文關鍵詞：	Eye-tracking, Neural-network, Inner Corner-Pupil Center Vector
相關次數：	點閱：12 下載：0
分享至:	分享至facebook 分享至twitter

查詢本校圖書館目錄查詢臺灣博碩士論文知識加值系統勘誤回報

眼睛是人類接收外界資訊的重要器官，同樣可以傳達資訊如注視的方向，周圍環境的明亮度等等，也是人類能夠用來表達情緒的器官，所以眼動追蹤一直都是熱門的研究課題。

近年來有越來越多的眼動追蹤儀推出市面，並且應用更為廣泛，心理學、醫學、教育、虛擬實境等都能看到眼動追蹤的應用。但目前大部分商用的眼動追蹤儀價格昂貴，並且頭部需要固定才能準確估測注視的方向。

本篇論文使用類神經網路結合本論文所提出的 Inner Corner-Pupil Center Vector (ICPCV)特徵，製作出可以讓使用者不用固定頭部而且不需要使用昂貴的硬體就能使用的眼動追蹤系統，並且與使用係數矩陣注視點估測演算法作比較，有更好的表現。

The eye is an important organ that human receives information via from the outside world. It also can convey information such as the direction of gaze, the
brightness of the surrounding environment and express emotions. Therefore, eye tracking has always been a popular research topic.

In recent years, there are more and more eye trackers produced and on sale.
The application of eye tracking is more wide-ranging such as psychology, medicine, education, virtual reality. But most of the eye tracker in the market is very expensive, and the head needs to be fixed in order to accurately estimate the direction of gaze.

This study develops an eye tracking system based on neural network and Inner Corner-Pupil Center Vector (ICPCV) feature which defined by ourselves.
Make it allows user move his/her head and needn’t expensive hardware. We compared with coefficient matrix gaze estimation algorithms and better than it.

摘要 i
ABSTRACT ii
誌謝 iii
目錄 iv
圖目錄 vi
表目錄 viii
第一章、 緒論 1
1-1 研究動機 1
1-2 研究目的 2
1-3 論文架構 3
第二章、 相關研究 4
2-1 眼動追蹤 4
2-2 眼動追蹤儀產品 5
2-3 眼動追蹤系統校正方法 7
2-4 注視點估測演算法 9
第三章、 研究方法 11
3-1 類神經網路 11
3-1-1 類神經網路簡介 11
3-1-2 類神經元 11
3-1-3 活化函數 12
3-1-4 正規化 13
3-2 感知機 13
3-2-1 感知機簡介 13
3-2-2 最小均方誤差演算法 14
3-3 二次方程式轉換 15
3-4 多層感知機 16
3-4-1 多層感知機簡介 16
3-4-2 倒傳遞演算法 17
3-5 放射狀基底函數網路 23
3-5-1 放射狀基底函數網路簡介 23
3-5-2 放射狀基底函數網路訓練學習法 25
3-6 深度神經網路 28
第四章、 實驗設計與步驟 29
4-1 實驗流程 29
4-2 資料收集 29
4-2-1 訓練資料 30
4-2-2 測試資料 31
4-3 擷取特徵 32
4-3-1 人臉擷取 32
4-3-2 眼部擷取 33
4-3-3 瞳孔中心擷取 36
4-3-4 眼角擷取 39
4-4 計算特徵值 40
4-4-1 Pupil Center - Eye Corner Vector 特徵 40
4-4-2 Inner Corner - Pupil Center Vector 特徵 42
4-5 資料過濾 44
4-6 訓練模型 44
4-6-1 類神經網路架構 44
4-6-2 計算係數矩陣 46
第五章、 實驗結果 47
5-1 類神經網路實驗結果 47
5-1-1 感知機實驗結果 47
5-1-2 二次方程式轉換實驗結果 49
5-1-3 多層感知機實驗結果 51
5-1-4 放射狀基底函數網路實驗結果 59
5-1-5 深度神經網路實驗結果 66
5-2 係數矩陣實驗結果 71
5-3 研究方法比較 71
5-3-1 頭部固定資料集 72
5-3-2 頭部移動資料集 74
5-4 眼動軌跡圖 77
5-4-1 頭部固定模型 78
5-4-2 頭部移動模型 83
5-5 眼動鍵盤 88
第六章、 結論與未來展望 90
6-1 結論 90
6-2 未來展望 91
參考文獻 92
                                

[1] The University of Bradford: Video Eye Tracker. [Online]. Available: ht
tp://www.bradford.ac.uk/research/rkt-centres/visual-computing/facilitie
s/eye-tracking. [Accessed: 02-Jun-2017].
[2] EYESO 眼動儀系统. [Online]. Available: http://www.eyeso.net. [Acce
ssed: 02-Jun-2017].
[3] SMI Eye Tracking Glasses. [Online]. Available: https://www.smivision.c
om/eye-tracking/product/eye-tracking-glasses. [Accessed: 02-Jun-2017].
[4] Tobii Pro Glasses 2. [Online]. Available: https://www.tobiipro.com/prod
uct-listing/tobii-pro-glasses-2. [Accessed: 02-Jun-2017].
[5] The Eye Tribe. [Online]. Available: https://s3.eu-central-1.amazonaws.c
om/theeyetribe.com/theeyetribe.com/dev/start/index.html#ui_tool. [Acc
essed: 02-Jun-2017].
[6] Tobii Pro X3-120. [Online]. Available: https://www.tobiipro.com/produ
ct-listing/tobii-pro-x3-120. [Accessed: 02-Jun-2017].
[7] 眼動儀有哪些品牌以及相應的價格範圍. [Online]. Available: https://k
knews.cc/zh-tw/tech/2a4mpez.html. [Accessed: 02-Jun-2017].
[8] The Eye Tribe UI evaluation window. [Online]. Available: https://s3.eucentral-1.amazonaws.com/theeyetribe.com/theeyetribe.com/dev/start/in
dex.html. [Accessed: 02-Jun-2017].
[9] Psychwire Eye-tracking Wiki. [Online]. Available: http://wiki.psychwir
e.co.uk/?page_id=187. [Accessed: 02-Jun-2017].
93
[10] H. R. Chennamma and X. Yuan, “A Survey on Eye-Gaze Tracking
Techniques,” Indian Journal of Computer Science and Engineering,
Vol. 4, No. 5, pp. 388-393, 2013.
[11] 蘇木春、張孝德編著，機器學習：類神經網路、模糊系統以及基因
演算法則，第二版，全華科技圖書，民國一百零一年。
[12] W. McCulloch and W. Pitts, “A logical calculus of the ideas immanent
in nervous activity,” Bulletin of Mathematical Biophysics Vol. 5, pp.
115–133, 1943.
[13] F. Rosenblatt, “The Perceptron--a perceiving and recognizing automaton,
” Cornell Aeronautical Laboratory, Report 85-460-1, 1957.
[14] D. E. Rumelhart, G. E. Hinton, and R.J. Williams, “Learning representati
ons by back-propagating errors,” Nature, Vol. 323, pp. 533-536, 1986.
[15] G. E. Hinton and R. R. Salakhutdinov, “Reducing the Dimensionality of
Data with Neural Networks,” Science, Vol. 313, Issue 5786, pp. 504-507,
2006.
[16] L. Sesma, A. Villanueva, and R. Cabeza, “Evaluation of Pupil CenterEye
Corner Vector for Gaze Estimation Using a Web Cam,” ETRA '12
Proceedings of the Symposium on Eye Tracking Research and
Applications, pp. 217-220, 2012.
[17] M. C. Su, Y. Z. Hsieh, C. H. Wang, and P. C. Wang, “A Jacobian Matrixbased
Learning Machine and Its Applications in Medical Diagnosis,”
IEEE Access, Vol. PP, Issue 99, 2017.
[18] 莊英杰，「追瞳系統之研發於身障者之人機介面應用」，國立中央大
學資訊工程研究所碩士論文，民國九十三年。
94
[19] 邱國鈞，「追瞳系統之研製及其應用」，國立中央大學資訊工程研究
所碩士論文，民國九十五年。
[20] J. Wang, G. Zhang, and J. Shi, “2D Gaze Estimation Based on PupilGlint
Vector Using an Artificial Neural Network,” Applied Sciences, Vol.
6, Issue 174, 2016.
[21] O. Ferhat, A. Llanza, and F. Vilariño, “A Feature-Based Gaze Estimation
Algorithm for Natural Light Scenarios,” Pattern Recognition and Image
Analysis. IbPRIA 2015. Lecture Notes in Computer Science, Vol. 9117,
pp. 569-576, 2015.
[22] Y. M. Cheung and Q. Peng, “Eye Gaze Tracking With a Web Camera in
a Desktop Environment,” IEEE Transactions on Human-Machine
Systems, Vol. 45, Issue 4, 2015.
[23] H. Cai, H.Yu, X. Zhou, and H. Liu, “Robust Gaze Estimation via
Normalized Iris Center-Eye Corner Vector,” International Conference
on Intelligent Robotics and Applications ICIRA 2016: Intelligent
Robotics and Applications, pp. 300-309, 2016.
[24] 眼球追蹤. [Online]. Available: http://vlsi2004.ee.ntu.edu.tw/belab/mid
term_oral_files/2011_100_1/100-1-mid-9.pdf. [Accessed: 18-Jun-2017].

簡易檢索 / 詳目顯示

相關論文