簡易檢索 / 詳目顯示

研究生: 方偉力
論文名稱: 以主成份分析法和線性鑑別分析法辨識想像左右手動
指導教授: 葉榮木
Yeh, Zong-Mu
蔡俊明
Tsai, Chun-Ming
學位類別: 碩士
Master
系所名稱: 機電工程學系
Department of Mechatronic Engineering
論文出版年: 2007
畢業學年度: 95
語文別: 中文
論文頁數: 60
中文關鍵詞: 大腦人機介面腦電波線性鑑別分析主成份分析法
英文關鍵詞: Brain Computer Interface (BCI), Electroencephalography (EEG), Linear Discriminant Analysis (LDA), Principal Component Analysis (PCA)
論文種類: 學術論文
相關次數: 點閱:254下載:34
分享至:
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
  • 腦電波是從人類的頭皮所量測到的訊號,經由放大、濾雜訊、特徵擷取、分類、辨識,藉由大腦人機介面,可用來控制義肢,或訓練人們的專注力。本研究是提出一個結合主成份分析法及線性鑑別分析法對腦電波訊號做特徵擷取,來提昇辨識想像左右手動的辨識率。本論文中的實驗共有四名受測者,年齡約在二十三歲到二十六歲之間,而實驗主要的內容,是讓受測者想像左手動及右手動並擷取腦電波訊號。在腦電波訊號處理的過程中,利用主成份分析法及線性鑑別分析法做特徵擷取,接著使用最近鄰居法則做資料分類,實驗結果顯示四位受測者的平均辨識率可達85.75%,此辨識率結果優於其他相關方法。

    Electroencephalogram (EEG) signals are recorded at the scalp surface through the electrodes. After amplifying, denoising, classifying and recognition the signal, they can be used to control machine equipment or training people for improving concentration by using a Brain Computer Interface(BCI) system. In this paper, a classification method based on Principal Component Analysis(PCA) and Linear Discriminant Analysis (LDA) is proposed to classify imagery tasks for left-hand and right-hand movements. Four healthy subjects, aged 23-26 years, were volunteered to participate in an experiment. During the experiment, two imagery tasks were to be stimuli. The feature extraction can be determined by the Principal Component (PCA) and Linear Discriminant Analysis (LDA). The method also uses Nearest Neighbor Rule (NNR) to classify the processed data. The experimental results show that the average accuracy rates is improved to 85.75%. According to the experimental results, the classification method based on Principal Component (PCA) and Linear Discriminant Analysis (LDA) is better than those of other literatures raised in this paper.

    致 謝...........I 摘 要...........II Abstract........III 目 錄... ........IV 圖目錄...........VI 表目錄.. ........VIII 第一章 緒 論... 1 1.1 研究背景與動機... 1 1.2 研究目的........ 3 1.3 大腦生理結構..... 4 1.4 研究架構........ 8 1.5 論文架構........ 9 第二章 文獻探討...........11 第三章 理論與方法......... 18 3.1 快速傅立葉轉換(Fast Fourier Transform, FFT)..... 18 3.2 主成份分析法(Principal Component Analysis, PCA).. 24 3.3 線性鑑別分析法(Linear discriminant Analysis, LDA).26 3.4 最近鄰居法則(Nearest Neighbor Rule, NNR)與一次一個辦識率(Leave One Out, LOO).......... ........28 3.5 二次分類器(Quadratic Classifier). 29 3.6 主成份分析法結合線性鑑別分析法..... 30 第四章 實驗設備與流程......................35 4.1 實驗設備.........................35 4.2 實驗設計與實驗流程........ ........36 第五章 實驗結果分析與討論.......... ........39 第六章 結論.............. ................60 參考文獻.................................61 圖目錄 圖 1.1 中樞神經運動控制系統圖.............2 圖 1.2 大腦人機介面應用圖 ................3 圖 1.3 各功能區所對應於大腦的位置[11]..... 5 圖 1.4 四種腦電波訊號圖[7].............. 8 圖 1.5 μ波受自主動作而產生抑制的訊號[1]... 8 圖 1.6 研究流程圖....... ................10 圖 2.1 ERD計算流程圖.....................13 圖 2.2 活動週期視窗的移動[2]..... ........14 圖 2.3 實驗流程圖....... ................15 圖 2.4 實驗流程圖....... ................16 圖 3.1 8點DIT FFT流程圖[27]..... ........20 圖 3.2 8點DIT FFT蝴蝶圖[27].............21 圖 3.3 DIT FFT蝴蝶處理單元[27]...........22 圖 3.4 DIF FFT蝴蝶圖[27]................23 圖 3.5 DIF FFT蝴蝶處理單元[27]...........23 圖 3.6 主成份分析法意示圖[30]............24 圖 3.7 線性鑑別分析法意示圖..............26 圖 3.8 WINE資料集的原始資料分佈圖.........30 圖 3.9 WINE資料集經PCA運算後的資料分佈圖.. 31 圖 3.10 WINE資料集經LDA運算後的資料分佈圖.. 31 圖 3.11 WINE資料集使用PCA+LDA後的資料佈圖. 32 圖 3.12 原始資料分佈圖....................33 圖 3.13 經PCA運算後的資料分佈圖............33 圖 3.14 經LDA運算後的資料分佈圖............34 圖 3.15 使用PCA+LDA後的資料佈圖...........34 圖 4.1 實驗設備.........................36 圖 4.2 International 10/20系統配置圖[40]....... 37 圖 4.3 實驗情怳.........................37 圖 4.4 剌激指令時序圖....................38 圖 5.1 去眼動雜訊指令框..................40 圖 5.2 選取眼動雜訊去除依據點之指令框..... 40 圖 5.3 原始腦電波訊號....................40 圖 5.4 去眼動雜訊後之腦電波訊號............41 圖 5.5 為未選取範圍之腦波訊號..............42 圖 5.6 選取完畢之腦電波訊號................42 圖 5.7 濾波後之腦電波訊號..................43 圖 5.8 Epochign Properties使用圖.........43 圖 5.9 每個電極的電壓波形圖................44 圖 5.10 輸出成數字檔.......................44 圖 5.11 輸入矩陣示意圖.....................45 圖 5.12 降維後矩陣示意圖....................45 圖 5.13 以時域腦電波訊號分析之實驗流程圖......47 圖 5.14 以頻域腦電波訊號分析之實驗流程圖......48 圖 5.15 受測者一之想像左手動的頻率對空間分佈的能量圖......... 50 圖 5.16 受測者一之想像右手動的頻率對空間分佈的能量圖..........50 圖 5.17 受測者二之想像左手動的頻率對空間分佈的能量圖......... 50 圖 5.18 受測者二之想像右手動的頻率對空間分佈的能量圖......... 51 圖 5.19 受測者三之想像左手動的頻率對空間分佈的能量圖......... 51 圖 5.20 受測者三之想像右手動的頻率對空間分佈的能量圖......... 51 圖 5.21 受測者四之想像左手動的頻率對空間分佈的能量圖......... 52 圖 5.22 受測者四之想像右手動的頻率對空間分佈的能量圖......... 52 圖 5.23 以二次分類法分析之結果直方圖....... 56 圖 5.24 以最近鄰居法分析之結果直方圖....... 57 圖 5.25 二次分類法與最近鄰居法之實驗結果比較圖...... 57 表目錄 表 1.1 大腦皮質各區的功能[11]....................7 表 2.1 常用之特徵擷取方法整理... ................11 表 2.2 常用之特徵轉換方法整理.....................12 表 2.3 不同想像種類數目的資訊傳輸率[24]...........15 表 2.4 文獻比較表...............................17 表 5.1 頻譜分析比較表.. ........................53 表 5.2 不同辨識方法處理腦電波資料之辨識率比較...... 55 表 5.3 本研究與其他文獻之比較.....................59

    [1]陳志瑋,「研究以小波神經網路作μ波即時鑑別」,國立成功大學機械工程學系碩士論文,2002。
    [2]吳岳昌,「探討ERD方法在腦機介面系統設計之效能」,國立交通大學電機與控制工程學系碩士論文,2003。
    [3]T. M. Vaughan, “Guest editorial brain-computer interface technology:
    A review of the second international meeting,” IEEE Trans. Rehabil.Eng., vol. 11, no. 2, pp. 94–109, 2003.
    [4]D. M. Taylor, S. I. H. Tillery, and A. B. Schwartz, “Direct cortical control of 3D neuroprosthetic devices,” Science, vol. 296, no. 5574, pp. 1829–1832, 2002.
    [5]謝維延,「禪定腦電波頻帶判讀系統設計」, 國立交通大學電機與控制工程學系碩士論文,2003
    [6]劉勇鈞,「應用於指動偵測之腦電波訊號分析系統」, 國立成功大學資訊工程學系 碩士論文,2004。
    [7]http://www.internetactu.net/wp-content/documents/bci_brainlab.jpg
    [8]B. Blankertz, G. Dornhege, M. Krauledat, K.-R.Müller, V. Kunzmann, F. Losch, and G. Curio, “The berlin brain-computer interface: EEG-based communication without subject training,” IEEE Trans. Rehabil. Eng., vol. 14, no. 2, pp. 147–152, 2006.
    [9]Xiang Liao, Dezhong Yao, Dan Wu, and Chaoyi Li, “Combining Spatial Filters for the Classification of Single-Trial EEG in a Finger Movement Task,” IEEE Trans. Biomed. Eng., vol. 54, no. 5, pp. 821-831, 2007.
    [10]http://www.dls.ym.edu.tw/neuroscience/nsdivide_c.htm
    [11]http://www.dls.ym.edu.tw/neuroscience/functional_c.htm
    [12]黃津操,「適應性類神經模糊推論系統辨識腦電波P300」,國立臺灣師範大學機電科技學系碩士論文,2005。
    [13]Baharan Kamousi, Zhongming Liu, and Bin He, Fellow, IEEE, “Classification of Motor Imagery Tasks for Brain-Computer Interface Applications by Means of Two Equivalent Dipoles Analysis,” IEEE Transactions On Neural Systems And Rehabilitation Engineering, vol. 13, no. 2, June 2005.
    [14]Byung G. Jo and Myung H. Sunwoo, “New Continuous-Flow Mixed-Radix (CFMR) FFT Processor Using Novel In-Place Strategy,” IEEE Transactions on Circuits and Systems-1:Regular Papers, vol. 52, no. 5, pp. 911-919, 2005.
    [15]McFarland D. J. McFarland and J. R. Wolpaw, “Sensorimotor rhythm-based brain computerinterface (BCI): Feature selection by regression improves performance,” IEEE Trans. Neural Syst. Rehabil. Eng., vol. 13, no. 3, pp. 372–379, 2005.
    [16]S. Lemm, B. Blankertz, G. Curio, and K.-R. Müller, “Spatio-spectral filters for improved classification of single trial EEG,” IEEE Trans. Biomed. Eng., vol.52, no. 9, pp. 1541–1548, 2005.
    [17]M. Fatourechi, A. Bashashati, R.Ward, and G. Birch, “A hybrid genetic algorithm approach for improving the performance of the LF-ASD brain computer interface” in Proc. Int. Conf. Acoust. Speech Signal Process., 2004, vol. 5, pp. 345–348.
    [18]D. J. McFarland and J. R.Wolpaw, “Sensorimotor rhythm-based brain computer
    interface (BCI): Feature selection by regression improves performance,” IEEE Trans. Neural Syst. Rehabil. Eng., vol. 13, no. 3, pp. 372–379, Sep. 2005.
    [19]B. Blankertz, G. Dornhege, C. Schaefer, R. Krepki, J. Kohlmorgen, K. Mueller, V. Kunzmann, F. Losch, and G. Curio, “Boosting bit rates and error detection for the classification of fast-paced motor commands based on single-trial EEG analysis,” IEEE Trans. Neural Syst. Rehabil. Eng., vol. 11, no. 2, pp. 127–131, Jun. 2003.
    [20]M. Kirby and C. Anderson, “Geometric analysis for the characterization of nonstationary time-series,” in Springer Applied Mathematical Sciences Series Celebratory Volume for the Occasion of the 70th Birthday of, L. Sirovich, E. Kaplan, J. Marsden, and K. R. K. Sreenivasan, Eds. New York: Springer-Verlag, 2003, ch. 8, pp. 263–292.
    [21]C. W. Anderson, E. A. Stolz, and S. Shamsunder, “Multivariate autoregressive models for classification of spontaneous electroencephalogram during mental tasks,” IEEE Trans. Biomed. Eng., vol. 45, no. 3, pp. 277–286, Mar. 1998.
    [22]G. Gage, K. Ludwig, K. Otto, E. Ionides, and D. Kipke, “Naive coadaptive cortical control,” J. Neural Eng., vol. 2, pp. 52–63, 2005. T. Rosipal, L. Trejo, and B. Matthews, “Kernel PLS-SVC for linear and nonlinear classification,” in Proc. 20th Int. Conf. Machine Learning(ICML-2003), Washington, DC, 2003, pp. 640–647.
    [23]Bernhard Obermaier, Christa Neuper, Christoph Guger, and Gert Pfurtscheller, “Information Transfer Rate in a Five-Classes Brain–Computer Interface,” IEEE Transactions On Neural Systems And Rehabilitation Engineering, vol. 9, no. 3, September 2001.
    [24]C. Guger, G. Edlinger, W. Harkam, I. Niedermayer, and G. Pfurtscheller, “How Many People are Able to Operate an EEG-Based Brain-Computer Interface (BCI)?”, IEEE Transactions On Neural Systems And Rehabilitation Engineering, VOL. 11, NO. 2, JUNE 2003.
    [25]Gert Pfurtscheller, Christa Neuper, Alois Schl¨ogl, and Klaus Lugger, “Separability of EEG Signals Recorded During Right and Left Motor Imagery Using Adaptive Autoregressive Parameters,” IEEE Transactions On Neural Systems And Rehabilitation Engineering, vol. 6, no. 3, September 1998.
    [26]沈世評,「用線性鑑別分析法做冥想的四個方向」,國立臺灣師範大學機電科技學系碩士論文,2005。
    [27]林志穎,「數位音訊廣播系統中轉換器之電路設計」,國立成功大學電機工程學系碩士論文,2001。
    [28]Aleix M. Martinez, and Avinash C. Kak, “PCA versus LDA,” IEEE Trans on Pattern Analysis and Machine Intelligence, vol.23, no.2, pp. 228 - 233 , 2001.
    [29]Jyh-Shing Roger Jang, "Data Clustering and Pattern Recognition," (in Chinese) available at the links for on-line courses at the author's homepage at http://www.cs.nthu.edu.tw/~jang.
    [30]張志豪、陳鴻彬、陳柏琳,「資料導向線性特徵轉換於中文大詞彙連續語音辨識之應用」, 「2005年全國計算機會議(National Computer Symposium, NCS)」, December 15-16, 2005。
    [31]Jieping Ye and Qi Li, “A Two-Stage Linear Discriminant Analysis via QR-Decomposition,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 27, pp. 929 – 941, 2005.
    [32]洪倩玉,「建立動態線性鑑別分析於線上人臉辨識與驗證」。國立成功大學資訊工程學系碩士論文,2003。
    [33]S. Noushath, Kumar G. Hemantha, and P. Shivakumara, “(2D)2LDA: An efficient approach for face recognition,” Pattern Recognition, vol 39, pp. 1396 – 1400, 2006.
    [34]陳若涵,「以音樂內容為基礎的情緒分析與辨識」。國立清華大學資訊與應用系統學系碩士論文,2006。
    [35]Keinosuke Fukunaga and Donald Hummels, “Leave-one-out Procedures for Nonparametric Error Estimates,” IEEE Trans on Pattern Analysis and Machine Intelligence, vol. 11, no. 4, pp. 421-423, 1989.
    [36]Ömer Nezih Gerek and Dogan Gökhan Ece, “Power-Quality Event Analysis Using Higher Order Cumulants and Quadratic Classifiers,” IEEE Transactions on Power Delivery, vol. 21, no. 2, pp. 883-889, 2006.
    [37]http://www.ics.uci.edu/~mlearn/MLRepository.html
    [38]http://www.neuroscan.com/landing.cfm
    [39]http://www.mathworks.com/
    [40]J. Malmivuo and R. Plonsey, Bioelectromagnetism, 1st ed. New York, New York: Oxford University Press, 1995.

    QR CODE