| 基于EMD和Hilbert变换的脑磁信号特征提取和分类 |
| |
| 引用本文： | 王健锋,黄晓霞.基于EMD和Hilbert变换的脑磁信号特征提取和分类[J].上海海事大学学报,2014,35(3):90-94. |
| |
| 作者姓名： | 王健锋 黄晓霞 |
| |
| 作者单位： | 上海海事大学信息工程学院,上海,201306 |
| |
| 摘 要： | 为研究通过脑机接口(Brain-Computer Interface,BCI)处理非线性、非稳定性信号问题,针对基于脑磁信号(magnetoencephalography,MEG)的BCI,提出一种基于经验模式分解(Empirical Mode Decomposition,EMD)和Hilbert变换的MEG特征提取和分类方法.该方法首先对MEG数据进行预处理;其次用EMD和Hilbert变换方法提取特征向量;然后用主成分分析法对提取到的特征向量进行降维处理;最后把处理过的特征向量作为支持向量机(Support Vector Machine,SVM)的一个输入向量实现MEG的分类.使用该方法对第4届国际BCI竞赛提供的MEG数据进行分类,实验结果表明可以获得较高的分类准确率.
|
| 关 键 词： | 脑机接口(BCI) 脑磁图(MEG) 经验模式分解(EMD) Hilbert变换 主成分分析 支持向量机(SVM) |
| 收稿时间： | 2013/10/17 0:00:00 |
| 修稿时间： | 3/5/2014 12:00:00 AM |
Feature extraction and classification of magnetoencephalography based on EMD and Hilbert transform |
| |
| WANG Jianfeng
,
HUANG Xiaoxia.Feature extraction and classification of magnetoencephalography based on EMD and Hilbert transform[J].Journal of Shanghai Maritime University,2014,35(3):90-94. |
| |
| Authors: | WANG Jianfeng HUANG Xiaoxia |
| |
| Institution: | Shanghai Maritime University,Information Engineering College and Shanghai Maritime University,College of Information Engineer |
| |
| Abstract: | In order to process nonlinear and non stationary signals by a Brain Computer Interface (BCI), a feature extraction and classification method of magnetoencephalography (MEG) based on Empirical Mode Decomposition (EMD) and Hilbert transform is proposed for an MEG based BCI. First, MEG data are preprocessed. Second, the feature vector is extracted by EMD and Hilbert transform, and then the dimension of the feature vector is reduced by the principal component analysis. Finally, the processed feature vector is used as an input vector of Support Vector Machine (SVM) to achieve the classification of MEG. The method is used to classify the MEG data of the BCI Competition IV, and the experimental result shows that the method is of high classification accuracy. |
| |
| Keywords: | Brain Computer Interface (BCI) magnetoencephalography (MEG) Empirical Mode Decomposition (EMD) Hilbert transform principal component analysis Support Vector Machine (SVM) |
| 本文献已被 CNKI 万方数据 等数据库收录！ |
| 点击此处可从《上海海事大学学报》浏览原始摘要信息 |
| 点击此处可从《上海海事大学学报》下载免费的PDF全文 |
