以主成份分析法和線性鑑別分析法辨識想像左右手動
No Thumbnail Available
Date
2007
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
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.
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.
Description
Keywords
大腦人機介面, 腦電波, 線性鑑別分析, 主成份分析法, Brain Computer Interface (BCI), Electroencephalography (EEG), Linear Discriminant Analysis (LDA), Principal Component Analysis (PCA)