首页|期刊简介|投稿指南|分类索引|刊文选读|订阅指南|资料|样刊邮寄查询|常见问题解答|联系我们
张学军,霍延,黄丽亚,等. 基于集合经验模式分解和公共空间模式的脑电信号特征提取[J]. 科学技术与工程, 2020, 20(1): 109-117.
Zhang Xuejun,HuoYan,Huang Liya,et al.A Novel EEG Feature Extraction based on EEMD and CSP[J].Science Technology and Engineering,2020,20(1):109-117.
基于集合经验模式分解和公共空间模式的脑电信号特征提取
A Novel EEG Feature Extraction based on EEMD and CSP
投稿时间:2019-04-25  修订日期:2019-09-22
DOI:
中文关键词:  集合经验模式分解 公共空间模式分解 FIR滤波器组 支持向量机
英文关键词:ensemble empirical mode decomposition common spatial pattern fir filter bank support vector machine
基金项目:国家自然科学基金
           
作者单位
张学军 南京邮电大学工程训练中心/电子与光学工程学院
霍延 南京邮电大学电子与光学工程学院
黄丽亚 南京邮电大学电子与光学工程学院
成谢锋 南京邮电大学电子与光学工程学院
摘要点击次数: 224
全文下载次数: 76
中文摘要:
      公共空间模式能够较好地提取运动想象任务时脑电信号的判别特性,但是其性能与大脑进行想象任务的频带相关。为了确定这样一组频带实现精确的分类,文章基于集合经验模式分解、FIR滤波器组以及公共空间模式算法提出了一种脑电特征提取方法。预处理去除伪迹后的信号首先经过集合经验模式算法获得多个模函数,然后选择出包含μ节律和β节律范围的分量实现信号重构,重构后的脑电信号作为带通滤波器组的输入得到若干子带信号集合,从每个子带信号中提取CSP特征,最后将提取的特征经过SVM进行分类。运用该方法对BCI竞赛数据集进行分类,实验表明该方法能够自适应地提取、筛选和判别每个受试者的空间CSP特征,分类准确率达96.53%。
英文摘要:
      The Common Spatial Pattern (CSP) was known to be effective in extracting feathers from Motor Imagery electroencephalograms. However its performance depended on the frequency bands that related to brain activities associated with motor Imagery tasks. In order to determine such a set of frequency bands to acquire an accurate classification, this paper proposed a novel EEG feather extraction method based on Ensemble Empirical Mode Decomposition (EEMD) and Common Spatial Pattern (CSP) combined with Finite Impulse Response (FIR) filter bank. This method can effectively perform the autonomous extraction and selection of key individual spatial discriminative CSP feather. The preprocessed EEG signal was decomposed into Intrinsic Mode Functions (IMFs) by EEMD. Then Intrinsic Mode Functions containing μ and β rhythms were selected to obtain reconstructed EEG signal, and reconstructed EEG signal was further decomposed into multiple sun-band signals by FIR filter bank. Subsequently, the feathers were extracted from each sun-band signal by CSP algorithm. Finally, Support Vector Machine (SVM) was used to classify CSP feather. This method was implemented on Brain-Computer Interface (BCI) competition data set, and the result reveal that proposed method obtained superior performance of as 96.53%.
查看全文  查看/发表评论  下载PDF阅读器
关闭
你是第28674581位访问者
版权所有:科学技术与工程编辑部
主管:中国科学技术协会    主办:中国技术经济学会
Tel:(010)62118920 E-mail:stae@vip.163.com
京ICP备05035734号-4
技术支持:本系统由北京勤云科技发展有限公司设计

京公网安备 11010802029091号