可選字及可再訓練眼寫系統｜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：	林鈺庭 Yu-Ting Lin
論文名稱：	可選字及可再訓練眼寫系統 Selectable and retrainable eye writing system
指導教授：	蔡章仁 Jang-Zern Tsai
口試委員:
學位類別：	碩士 Master
系所名稱：	資訊電機學院 - 電機工程學系 Department of Electrical Engineering
論文出版年：	2021
畢業學年度：	109
語文別：	中文
論文頁數：	61
中文關鍵詞：	眼電圖、眼寫、深度學習、卷積網路、數據擴增
外文關鍵詞：	EOG, eye-writing, deep learning, CNN, data augmentation
相關次數：	點閱：15 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

患有肌萎縮側索硬化症 (Amyotrophic lateral sclerosis, ALS)的病人，由
於運動神經元的退化，肌肉會逐漸萎縮，失去運動能力，無法利用口說和肢
體對外溝通，較晚開始退化的眼球運動功能成為後期病人希望可以依靠來
對外溝通的一個途徑。
本研究以阿拉伯數字(0~9)、大寫英文字母(A~Z)以及 4 個特殊符號(空
格、輸入、句點、問號)，共 40 個符號，建立一套眼寫系統。使用眼電圖法
（Electrooculography, EOG）記錄眼球運動，將 EOG 訊號轉為影像利用卷積
神經網路(Convolutional Neural Networks，CNN）進行符號辨識，以眨眼數
次當作特殊指令來控制進出書寫辨識次系統、書寫的開始與結束、刪字或選
字，以模擬一般的手寫和打字情境。
模型訓練所需要的資料包含全部符號各被眼寫多次所收集的原始資料
以及擴增資料，以 k 折交叉驗證（k-fold cross-validation）的方式訓練卷積
網路。使用者實際使用該系統時，可以辨識或選擇成功的符號軌跡可用來再
訓練，使模型更加穩健，越使用越符合使用者的書寫習慣，提高眼寫辨識的
準確率(Accuracy)。

In patients with amyotrophic lateral sclerosis (ALS), due to the degeneration
of motor neurons, the muscles will gradually atrophy, lose exercise ability, and
cannot use oral and physical communication. Eye moving function that
degenerates later may become the only way for the later-stage patients to rely on
to communicate with others.
This study uses Arabic numerals (0~9), uppercase English letters (A~Z) and
4 special symbols (space, input, dot, question mark), a total of 40 symbols, to
establish the eye-writing system. This system uses electrooculography (EOG) to
record eye movements. It converts the EOG signals into images for symbol
recognition with a convolutional neural network (CNN). It uses the number of
blinks as special commands to control the entering and exiting of the writing and
recognition subsystem, the start and end of writing, character selection and
deletion. The system functions to emulate general handwriting and typing
situations.
The data required for model training include the original data collected by
eye-writing all symbols for multiple times and the augmented data. The CNN is
trained by k-fold cross-validation. When the system is practically used by the user,
any symbol trace that leads to successful character recognition or selection can be
used to retrain the neural network model to make the model gradually become
more robust. The more the system is used and retrained, the more it will fit the
user's writing habits, and the accuracy of eye-writing recognition will be gradually
improved.

摘要........................................................................................................................i
Abstract................................................................................................................ii
目錄......................................................................................................................iii
圖目錄.................................................................................................................. vi
表目錄................................................................................................................viii
一、 緒論........................................................................................................... 1
1-1 研究動機................................................................................................... 1
1-2 論文架構................................................................................................... 1
1-3 文獻回顧................................................................................................... 2
二、 背景知識................................................................................................... 4
2-1 EOG 訊號 ................................................................................................... 4
2-2 其他眼動量測方法 ................................................................................... 6
2-3 眼動應用.................................................................................................... 7
三、 眼寫系統介紹........................................................................................... 8
iv
3-1 硬體設備.................................................................................................... 8
3-1-1 電極貼片 ............................................................................................ 8
3-1-2 儀表放大器 ........................................................................................ 9
3-1-3 右腳驅動電路.................................................................................. 10
3-1-4 高通濾波器 ...................................................................................... 11
3-1-6 低通濾波器 ...................................................................................... 12
3-1-5 反向加法器 ...................................................................................... 12
3-1-7 類比數位轉換器.............................................................................. 14
3-3 眨眼偵測.................................................................................................. 15
3-2 訊號處理.................................................................................................. 15
3-4 資料擴增.................................................................................................. 19
3-5 網路結構.................................................................................................. 22
3-6 系統操作.................................................................................................. 24
四、 實驗設置................................................................................................. 28
4-1 符號集...................................................................................................... 28
4-2 實驗人員.................................................................................................. 31
4-3 實驗項目.................................................................................................. 31
4-3-1 個人化系統 ...................................................................................... 31
v
4-3-2 資料擴增 .......................................................................................... 32
4-3-3 可選字及可再訓練.......................................................................... 32
五、 結果與討論............................................................................................. 33
5-1 個人系統.................................................................................................. 33
5-1-1 準確率 .............................................................................................. 33
5-1-2 混淆矩陣 .......................................................................................... 35
5-1-3 可靠度與可信度.............................................................................. 37
5-2 資料擴增.................................................................................................. 40
5-3 可選字及可再訓練 ................................................................................. 42
5-3-1 可選字 .............................................................................................. 42
5-3-2 再訓練 .............................................................................................. 42
5-4 與其他眼寫系統的比較 ......................................................................... 43
六、 結論與未來展望..................................................................................... 45
6-1 結論.......................................................................................................... 45
6-2 未來展望.................................................................................................. 46
參考文獻............................................................................................................. 48

                                

[1] Rowland LP, Shneider NA, “Amyotrophic lateral sclerosis”, N Engl J Med,
Vol. 344, No. 22, pp.1688-1700, 2001 May 31.
[2] EFNS Task Force on Diagnosis and Management of Amyotrophic Lateral
Sclerosis:, Andersen PM, Abrahams S, Borasio GD, de Carvalho M, Chio A,
Van Damme P, Hardiman O, Kollewe K, Morrison KE, Petri S, Pradat PF,
Silani V, Tomik B, Wasner M, Weber M, “EFNS guidelines on the clinical
management of amyotrophic lateral sclerosis (MALS)--revised report of an
EFNS task force”, Eur J Neurol, Vol. 19, No. 3, pp. 360-375, 2012 Mar.
[3] Sharma R, Hicks S, Berna CM, Kennard C, Talbot K, Turner MR,
“Oculomotor dysfunction in amyotrophic lateral sclerosis: a comprehensive
review”, Arch Neurol, Vol. 68, No. 7, pp. 857-861, 2011 Jul.
[4] T. E. Hutchinson, K. P. White, W. N. Martin, K. C. Reichert and L. A. Frey,
“Human-computer interaction using eye-gaze input”, IEEE Transactions on
Systems, Man, and Cybernetics, Vol. 19, no. 6, pp. 1527-1534, 1989 Dec.
[5] A. E. Kaufman, A. Bandopadhay and B. D. Shaviv, “An eye tracking
computer user interface”, Proceedings of 1993 IEEE Research Properties in
Virtual Reality Symposium, pp. 120-121, 1993 Oct.
[6] Y. Tomita, Y. Igarashi, S. Honda and N. Matsuo, “Electro-oculography
mouse for amyotrophic lateral sclerosis patients”, Proceedings of 18th
Annual International Conference of the IEEE Engineering in Medicine and
Biology Society, Vol.5, pp. 1780-1781, 1996 Oct.
[7] Tsai, J. Z., Lee, C. K., Wu, C. M., Wu, J. J., and Kao, K. P, “A Feasibility
Study of an Eye-writing System Basedon Electro-oculography”, Journal of
49
Medical and Biological Engineering, Vol. 28, No. 1, pp. 39-46, 2008 Mar.
[8] Diard J, Rynik V, Lorenceau J, “A Bayesian computational model for online
character recognition and disability assessment during cursive eye writing”,
Front Psychol, Vol. 4, No. 843, 2013 Nov 11.
[9] Lorenceau J, “Cursive writing with smooth pursuit eye movements”, Curr
Biol, Vol. 22, No. 16, pp. 1506-1509 , 2012 Aug.
[10] K. Lee, W. Chang, S. Kim and C. Im, “Real-Time “Eye-Writing”
Recognition Using Electrooculogram”, IEEE Transactions on Neural
Systems and Rehabilitation Engineering, Vol. 25, No. 1, pp. 37-48, 2017
Jan.
[11] Fang F, Shinozaki T, “Electrooculography-based continuous eye-writing
recognition system for efficient assistive communication systems”, PLoS
One, Vol. 19 , No. 2, 2018 Feb 9.
[12] Fang F, Shinozaki T, Horiuchi Y, Kuroiwa S, Furui S, Musha T, “Improving
Eye Motion Sequence Recognition Using Electrooculography Based on
Context-Dependent HMM”, Comput Intell Neurosci, 2016.
[13] M. Merino, O. Rivera, I. Gómez, A. Molina and E. Dorronzoro, “A Method
of EOG Signal Processing to Detect the Direction of Eye Movements”,
2010 First International Conference on Sensor Device Technologies and
Applications, pp. 100-105, 2010.
[14] Marmor MF, Brigell MG, McCulloch DL, Westall CA, Bach M;
International Society for Clinical Electrophysiology of Vision, “ISCEV
standard for clinical electro-oculography (2010 update)”, Doc Ophthalmol,
Vol. 122, No. 1, pp. 1-7, 2011 Feb.
[15] Duchowski A, Eye Tracking Methodology, Eye Tracking Techniques,
Springer, London, 2007.
50
[16] D. A. Robinson, “A Method of Measuring Eye Movemnent Using a Scieral
Search Coil in a Magnetic Field”, IEEE Transactions on Bio-medical
Electronics, Vol. 10, No. 4, pp. 137-145, 1963 Oct.
[17] Gans, Richard E, “Video-oculography”, The Hearing Journal, Vol. 54, No.
5, pp. 40-42, 2001 May.
[18] C. Buquet, J. Charlier, G. Dhélin, S. Toucas, M. Quéré and I. Moati,
“Photo-oculography: A new method for eye movements study, interest in
ophthalmological and extra pyramidal neurological diseases”, 1992 14th
Annual International Conference of the IEEE Engineering in Medicine and
Biology Society, pp. 1555-1556, 1992.
[19] Clark R, Blundell J, Dunn MJ, Erichsen JT, Giardini ME, Gottlob I, Harris
C, Lee H, Mcilreavy L, Olson A, Self JE, Vinuela-Navarro V, Waddington
J, Woodhouse JM, Gilchrist ID, Williams C, “The potential and value of
objective eye tracking in the ophthalmology clinic”, Eye (Lond), Vol. 33,
No. 8, pp. 1200-1202, 2019 Aug.
[20] Papageorgiou E, McLean RJ, Gottlob I. “Nystagmus in childhood”, Pediatr
Neonatol, Vol. 55, No. 5, pp. 341-351, 2014 Oct.
[21] Armstrong RA, “Oculo-Visual Dysfunction in Parkinson's Disease”, J
Parkinsons Dis, Vol. 5, No. 4, pp. 715-726, 2015.
[22] Benson PJ, Beedie SA, Shephard E, Giegling I, Rujescu D, St Clair D,
“ Simple viewing tests can detect eye movement abnormalities that
distinguish schizophrenia cases from controls with exceptional accuracy”,
Biol Psychiatry, Vol. 72, No. 9, pp. 716-24, 2012 Nov 1.
[23] Barreto, A. M. , “Do users look at banner ads on facebook?”, Journal of
Research in Interactive Marketing, Vol. 7, No. 2, pp. 119-139, 2013.
[24] Joseph H Goldberg, Xerxes P Kotval, “Computer interface evaluation using
51
eye movements: methods and constructs”, International Journal of
Industrial Ergonomics, Vol.24, No.6, pp. 931-645, 1999 Oct.
[25] 維基百科 : 儀表放大器。2021 年 5 月 1 日，取自
https://zh.wikipedia.org/wiki/%E5%84%80%E8%A1%A8%E6%94%BE%
E5%A4%A7%E5%99%A8。
[26] INA128P Datasheet(PDF) 1 Page - Burr-Brown (TI)。2021 年 5 月 1
日，取自 https://html.alldatasheet.com/html-pdf/56684/BURRBROWN/INA128P/501/1/INA128P.html。
[27] S. Albawi, T. A. Mohammed and S. Al-Zawi, “Understanding of a
convolutional neural network”, 2017 International Conference on
Engineering and Technology (ICET), pp. 1-6, 2017.

簡易檢索 / 詳目顯示

相關論文