臺灣博碩士論文加值系統

在這網路快速發展的時代，用戶的資料安全成為了一個相當重要的問題。起
初都是透過一組密碼來存取使用者資料，若使用複雜的密碼，可以保有較高的安
全性，但相對的使用者較不方便使用。然而使用過度簡單的密碼較容易遭到有心
人士破解，因此在後來加入各種不同的驗證碼，用來識別使用者是否為真人。而
我們將在驗證碼輸入的過程當中來辨別使用者是否為本人，達到免除密碼的效
果。
我們將透將過使用者本身所具備的生物特徵，做為識別的依據。在本篇論文
中，將使用腦電圖(EEG) 的生物識別系統(Biometric System)。我們提出一個反
向傳播神經網路架構，針對每一位使用者，我們將透過使用者資料與其他使用者
資料比例的不同來訓練神經網路，經由交叉驗證選出辨識效果最佳的兩個神經網
路組合，並由該組合的輸出值做為我們辨識的判別依據。我們使用20 筆使用者資
料做訓練能達到平均94.05% 的辨識率，在只有10 筆資料訓練的情況下的辨識率
也能達到92.36%。而我們也使用我們的資料去比較其他做法的辨識能力，如一般
神經網路(Neural Network)(平均80.81%)，SVM with 10 cross validation(72.35%)。
我們的做法能在相同資料下，比起其他方法提升約13 至21% 左右的辨識效果。
我們提升了辨識的準確度，並改善錯誤接收率以及錯誤拒絕率。另外還能減少使
用者所需的訓練資料數，以及資料輸入時間。
關鍵字：EEG，視覺驗證碼，身分認證。

With the rapid development of internet, keeping the security of user accounts
has become essential. Initially, People set fixed password to access the data. Using
the complex password have a higher security, but not convenient relatively. It will
be easier to break the password by people with bad intention, if using too easy
password. CAPTCHA was then added to identify the actual human by different
types of architectures. We will identify the user when entry the CAPTCHA, that
can achieve free password.
The biological characteristic of the users can be adopted, as the basis for identification.
In this paper, biometric system of EEG was used. We propose a back
propagation neural network architecture. For every user, we will use different data
proportion of user and others to train the network. We choose the best identification
performance two network combination by using cross validation, and the combination’s
output for our identification. The identification rate achieve 94.05% when
20 user’s training data be used, and it reached 92.36% when only 10 training data
be used. We also used our data to compare with other methods, such as neural
network with the average of 80.81% and SVM with 10 cross validation 73.25%. Our
method can enhance about 13-21% compared to other methods in the same data.
We improve the accuracy of identification, reduce the false accept rate and false
reject rate. Also reduce the number of training data and data entry time needed by
the user.
Keyword：EEG, Visual CAPTCHA, Identity Authentication.

論文指導教授推薦書. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . i
考試委員審定書. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ii
中文摘要. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iii
英文摘要. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . iv
誌謝. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . v
目錄. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vi
表目錄. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii
圖目錄. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix
1 簡介. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 背景. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 研究動機與目的. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.3 研究目的. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.4 論文架構. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2 相關研究. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.1 EEG 的發展沿革. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.2 靜止狀態下的身份辨識. . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.3 外部刺激的身份辨識. . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.4 驗證碼. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3 方法介紹. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3.1 大腦與裝置介紹. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3.2 資料擷取的情境介紹與驗證碼. . . . . . . . . . . . . . . . . . . . . . 13
3.2.1 情境介紹. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
3.2.2 驗證碼的設計與使用. . . . . . . . . . . . . . . . . . . . . . . 15
3.3 濾波分解腦電圖訊號. . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.4 特徵擷取. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
3.5 參數設定. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
3.5.1 隱藏層的層數. . . . . . . . . . . . . . . . . . . . . . . . . . . 20
3.5.2 隱藏層節點數. . . . . . . . . . . . . . . . . . . . . . . . . . . 21
3.5.3 誤差容忍值. . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
3.5.4 訓練次數. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.6 神經網路架構. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
3.6.1 訓練資料. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25
3.6.2 測試資料. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
4 實驗與效能評估. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
4.1 實驗評估方式. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
4.2 實驗環境. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
4.3 實驗分析. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
4.3.1 與其他單節點的方法比較. . . . . . . . . . . . . . . . . . . . 38
5 結論與未來展望. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
參考文獻. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41
附錄甲：神經網路測試結果. . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

[1] W. Khalifa, A. Salem, M. Roushdy, and K. Revett, “A survey of EEG based user authentication schemes,” in 2012 8th International Conference on Informatics and Systems (INFOS), pp. BIO–55–60, 2012.
[2] A. Almehmadi and K. El-Khatib, “The state of the art in electroencephalogram and access control,” in 2013 Third International Conference on Communications and Information Technology (ICCIT), pp. 49–54, IEEE, 2013.
[3] J. Thorpe, P. C. van Oorschot, and A. Somayaji, “Pass-thoughts: authenticating with our minds,” in Proceedings of the 2005 workshop on New Security Paradigms, pp. 45–56, ACM, 2005.
[4] D. Peralta, I. Triguero, R. Sanchez-Reillo, F. Herrera, and J. M. Benítez, “Fast fingerprint identification for large databases,” Pattern Recognition, vol. 47, no. 2, pp. 588–602, 2014.
[5] S. Pankanti, S. Prabhakar, and A. K. Jain, “On the individuality of fingerprints,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 24, no. 8, pp. 1010–1025, 2002.
[6] Y. Xu, L. Fei, and D. Zhang, “Combining left and right palmprint images for more accurate personal identification,” IEEE Transactions on Image Processing, vol. 24, no. 2, pp. 549–559, 2015.
[7] M. Blanton and P. Gasti, “Secure and efficient protocols for iris and fingerprint identification,” in European Symposium on Research in Computer Security, pp. 190–209, Springer, 2011.
[8] A. Dustor and P. Kłosowski, “Biometric voice identification based on fuzzy kernel classifier,” in International Conference on Computer Networks, pp. 456–465, Springer, 2013.
[9] Y. Sun, Y. Chen, X. Wang, and X. Tang, “Deep learning face representation by joint identification-verification,” in Advances in Neural Information Processing Systems, pp. 1988–1996, 2014.
[10] S. Liu, Y. Bai, J. Liu, H. Qi, P. Li, X. Zhao, P. Zhou, L. Zhang, B. Wan, C. Wang, et al., “Individual feature extraction and identification on EEG signals in relax and visual evoked tasks,” Biomedical Informatics and Technology, pp. 305–318, 2014.
[11] S. Marcel and J. d. R. Millán, “Person authentication using brainwaves EEG and maximum a posteriori model adaptation,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 29, no. 4, 2007.
[12] M. Poulos, M. Rangoussi, and N. Alexandris, “Neural network based person identification using EEG features,” in 1999 IEEE International Conference on Acoustics, Speech, and Signal Processing, vol. 2, pp. 1117–1120, IEEE, 1999.
[13] M. Poulos, M. Rangoussi, V. Chrissikopoulos, and A. Evangelou, “Person identification based on parametric processing of the EEG,” in The 6th IEEE International Conference on Electronics, Circuits and Systems. (ICECS’99), vol. 1, pp. 283–286, IEEE, 1999.
[14] M. Abo-Zahhad, S. M. Ahmed, and S. N. Abbas, “A novel biometric approach for human identification and verification using eye blinking signal.,” 2015 IEEE Signal Process Letters, vol. 22, no. 7, pp. 876–880, 2015.
[15] L. F. Haas, “Hans Berger(1873–1941), Richard Caton(1842–1926), and electroencephalography,” Journal of Neurology, Neurosurgery & Psychiatry, vol. 74, no. 1, pp. 9–9, 2003.
[16] M. Teplan et al., “Fundamentals of EEG measurement,” Measurement Science Review, vol. 2, no. 2, pp. 1–11, 2002.
[17] M. Poulos, M. Rangoussi, V. Chrissikopoulos, and A. Evangelou, “Parametric person identification from the EEG using computational geometry,” in The 6th IEEE International Conference on Electronics, Circuits and Systems, Proceedings of ICECS’99, vol. 2, pp. 1005–1008, IEEE, 1999.
[18] G. Mohammadi, P. Shoushtari, B. Molaee Ardekani, and M. B. Shamsollahi, “Person identification by using ar model for EEG signals,” in Proceeding of World Academy of Science, Engineering and Technology, vol. 11, pp. 281–285, 2006.
[19] H. A. Shedeed, “A new method for person identification in a biometric security system based on brain EEG signal processing,” in World Congress on Information and Communication Technologies (WICT), pp. 1205–1210, IEEE, 2011.
[20] T. Pham, W. Ma, D. Tran, P. Nguyen, and D. Phung, “EEG-based user authentication using artifacts,” in International Joint Conference SOCO' 14-CISIS' 14-ICEUTE' 14. Springer, pp. 343–353, Springer, 2014.
[21] Q. Gui, Z. Jin, and W. Xu, “Exploring EEG-based biometrics for user identification and authentication,” in IEEE Signal Processing in Medicine and Biology Symposium (SPMB), pp. 1–6, IEEE, 2014.
[22] A. B. Jeng, C.-C. Tseng, D.-F. Tseng, and J.-C. Wang, “A study of CAPTCHA and its application to user authentication,” in International Conference on Computational Collective Intelligence, pp. 433–440, Springer, 2010.
[23] E. Bursztein, M. Martin, and J. Mitchell, “Text-based CAPTCHA strengths and weakness,” in Proceedings of the 18th ACM conference on Computer and communications security, pp. 125–138, ACM, 2011.
[24] E. E. Papalexakis, A. Fyshe, N. D. Sidiropoulos, P. P. Talukdar, T. M. Mitchell, and C. Faloutsos, “Good-enough brain model: Challenges, algorithms, and discoveries in multisubject experiments,” 2014 Big data, vol. 2, no. 4, pp. 216–229, 2014.
[25] A. Prochazka, J. Kukal, and O. Vysata, “Wavelet transform use for feature extraction and EEG signal segments classification,” in 2008. 3rd International Symposium on Communications, Control and Signal Processing,(ISCCSP), pp. 719–722, IEEE, 2008.
[26] S. Karsoliya, “Approximating number of hidden layer neurons in multiple hidden layer BPNN architecture,” International Journal of Engineering Trends and Technology, vol. 3, no. 6, pp. 714–717, 2012.
[27] M. D. Odom and R. Sharda, “A neural network model for bankruptcy prediction,” in International Joint Conference on Neural Networks,(IJCNN), pp. 163–168, IEEE, 1990.
[28] S.-K. Yeom, H.-I. Suk, and S.-W. Lee, “Person authentication from neural activity of face-specific visual self-representation,” Pattern Recognition, vol. 46, no. 4, pp. 1159–1169, 2013.