|本期目录/Table of Contents|

[1]高诺△,高志栋,张慧,等.运动想象脑电信号特征提取与分类的黎曼方法研究*[J].生物医学工程研究,2021,03:246-251.
 GAO Nuo,GAO Zhidong,ZHANG Hui,et al.Riemannian approach research for the feature extraction and classification of motor imagery electroencephalogram(EEG) signals[J].Journal of Biomedical Engineering Research,2021,03:246-251.
点击复制

运动想象脑电信号特征提取与分类的黎曼方法研究*(PDF)

《生物医学工程研究》[ISSN:1006-6977/CN:61-1281/TN]

期数:
2021年03期
页码:
246-251
栏目:
出版日期:
2021-09-25

文章信息/Info

Title:
Riemannian approach research for the feature extraction and classification of motor imagery electroencephalogram(EEG) signals
文章编号:
1672-6278 (2021)03-0246-06
作者:
高诺△高志栋张慧陈鹏程
山东建筑大学信息与电气工程学院,济南 250101
Author(s):
GAO Nuo GAO Zhidong ZHANG Hui CHEN Pengcheng
Information & Electrical Engineering Department, Shandong Jianzhu University, Jinan 250101, China
关键词:
脑机接口运动想象脑电信号功率谱密度矩阵黎曼距离分类算法
Keywords:
Brain-computer interface Motor imageryElectroencephalogram Power spectral density matrix Riemannian distance Classification algorithm
分类号:
R318
DOI:
DOI10.19529/j.cnki.1672-6278.2021.03.04
文献标识码:
A
摘要:
针对目前脑电信号分析与处理算法尚无法满足脑机接口技术的应用要求,本研究提出了一种基于黎曼空间的运动想象脑电信号特征提取与分类方法。该方法以脑电信号的功率谱密度矩阵作为特征,以黎曼距离作为不同种类脑电信号之间相似性/非相似性度量,使用K最近邻算法作为最终的分类方法,对不同运动想象脑电信号进行黎曼空间的分析与处理。为了验证该方法的有效性与可行性,本研究进行了一系列相关实验。实验结果说明,本研究方法可以在样本集数量较小的情况下,对运动想象脑电信号进行分类,分类方法具有较高的分类准确率、较小的计算量和较快的计算速度,体现了基于黎曼空间的脑电信号分析与处理方法的优越性。
Abstract:
In view of the current EEG signal analysis and processing algorithm cannot meet the application requirements of brain-computer interface (BCI) technology, we proposed a feature extraction and classification method of motor imagery EEG signal based on Riemannian space.The power spectral density (PSD) matrix of the EEG signal was used as the feature, the Riemannian distance was used as the similarity/dissimilarity measure between different classes of EEG signals, and K-nearest neighbor (KNN) algorithm was adopted as the final classification method. In order to verify the feasibility and effectiveness of this method, a series of experiments were conducted. The experimental results indicate that this method can classify the motor imagery EEG signals with high classification accuracy, small amount of calculation and fast calculation speed, which reflects the superiority of the EEG signal analysis and processing method based on Riemannian space.

参考文献/References

备注/Memo

备注/Memo:
(收稿日期:2020-12-14)山东省重点研发计划项目(2017CXGC1505)。△通信作者Email: 1551170192@qq.com
更新日期/Last Update: 2021-10-26