基於腦波相關係數方法提取有效想像運動腦波;Effective MI Brain Wave Extraction based on Correlation Coefficient Method

NCUIR > College of Electrical Engineering & Computer Science > Graduate Institute of Electrical Engineering > Electronic Thesis & Dissertation > Item 987654321/84270

Please use this identifier to cite or link to this item: http://ir.lib.ncu.edu.tw/handle/987654321/84270

Title:	基於腦波相關係數方法提取有效想像運動腦波;Effective MI Brain Wave Extraction based on Correlation Coefficient Method
Authors:	藍子鈞 Lan, Zih-Jyun
Contributors:	電機工程學系
Keywords:	腦電圖;腦機介面;想像運動;相關係數;共同空間形樣法;支持向量機;虛擬實境;Electroencephalography;Brain-computer interface;Motor imagery;Correlation coefficient;Common spatial patterns;Support vector machine;Virtual reality
Date:	2020-07-29
Issue Date:	2020-09-02 18:46:10 (UTC+8)
Publisher:	國立中央大學
Abstract:	本論文基於時間序列計算相關係數，用於偵測想像運動(Motor Imagery, MI)開始時間，擷取有效的腦電圖(Electroencephalography, EEG)訊號，提出方法利用??波的相關係數，能夠找到有效EEG 訊號的起始時間點。因而降低計算負擔並有效減少共空間形樣法(Common Spatial Patterns, CSP)特徵提取的資料量大小，最後使用支持向量機(Support Vector Machine, SVM)達成分類準確度的提升。此外，提出方法在腦機介面(Brain-Computer Interface, BCI)的應用上，結合虛擬實境(Virtual Reality, VR)提出偵測想像運動的演算法。;The thesis, based on time series, calculates the correlation coefficient, and then detects the start time of motor imagery (MI). Moreover, the thesis proposes a method to capture the effective Electroencephalography (EEG) signal. Then using the correlation coefficient of the ?? wave, the method could find out the starting position of the effective EEG signal. Therefore, it dramatically reduces the amount of EEG data, which effectively reduces the computation load for feature extracted by common spatial patterns (CSP). Finally support vector machine (SVM) is used to improve the classification accuracy. Furthermore, in the application of brain-computer interface (BCI) combined with virtual reality (VR), an algorithm for detecting MI is demonstrated.
Appears in Collections:	[Graduate Institute of Electrical Engineering] Electronic Thesis & Dissertation

Files in This Item:

File	Description	Size	Format
index.html		0Kb	HTML	133	View/Open

View Licence

社群 sharing

Loading...