摘要
脑-机接口(brain-computer interface,BCI)为无法进行交流的人们提供了一种新的交流方式。传统的基于频率特征的脑电信号(electroencephalogram,EEG)特征提取方法只提取每个通道的能量特征,而忽略了不同通道之间的相关性信息。为了获得更好的特征提取结果,本研究采用了基于小波包和共同空间模型(common space pattern,CSP)的脑电信号特征提取方法。首先,在利用小波包对脑电信号分解前,对相关通道和频带进行辨别,提取运动想象脑电μ律和β节律,然后利用CSP算法进行空间滤波提取特征,选取相关节点计算小波包能量,最后通过支持向量机(support vector machine,SVM)将脑电信号分为左右手两种特征。为了验证本研究算法的可行性与有效性,在BCI竞赛数据集上进行了相应的实验,分类结果表明,所提出的特征提取算法能够有效提取运动想象特征,具有较高的分类精度。
Brain-computer interface(BCI)provides a new way for people who cannot communicate.The traditional EEG feature extraction method based on frequency feature only extracts the energy features of each channel,but ignores the correlation information between different channels.In order to obtain better feature extraction results,we adopted the method of EEG signal feature extraction based on wavelet packet and common space pattern(CSP).First,before the use of wavelet packet decomposition of EEG signals,the relevant channel and frequency band were identified,the movement imagine electrical rhythm was extracted including μ rhythm and β rhythm,then the features were extracted using CSP spatial filtering algorithm,the node calculation of wavelet packet energy was selected,finally,the brain electrical signals were divided into left and right hand characteristics with support vector machine(SVM).In order to verify the feasibility and effectiveness of the proposed algorithm,we carried out corresponding experiments on the BCI competition data set.The results show that the proposed feature extraction algorithm can effectively extract motion imagination features and has high classification accuracy.
作者
高枫
鲁昊
高诺
GAO Feng;LU Hao;GAO Nuo(Information&Electrical Engineering Department,Shandong Jianzhu University,Jinan 250101,China)
出处
《生物医学工程研究》
2019年第4期393-396,409,共5页
Journal Of Biomedical Engineering Research
基金
全国大学生创新创业计划项目(0001222)
山东省重点研发计划项目(2017CXGC1505)
2018年山东建筑大学教学建设与改革重点项目(010171820)
关键词
小波包分析
共同空间模型
支持向量机
脑机接口
运动想象
特征提取
Wavelet packet analysis
Common space pattern
Support vector machines
Brain-computer interface
Motor Imagery
Feature extraction