基于变分模态分解与深度信念网络的运动想象分类识别研究

doi:10.3969/j.issn.1000-1158.2020.01.17

摘要
图/表
参考文献(27)
相关文章 (15)

全文: PDF (2068 KB) HTML (1 KB)
输出: BibTeX | EndNote (RIS)

摘要传统人工确定最优时段及最优频段的方法会造成信息遗漏进而导致运动想象识别率的降低, 因此基于脑电信号的运动想象分类研究成为了脑-机接口研究领域的难点问题。针对该问题,变分模态分解和深度信念网络被应用于运动想象分类。对脑电信号进行变分模态分解得到窄带分量,利用希尔伯特变换提取边际谱、特征频带下的瞬时能谱以及时-频联合特征; 特征融合后采用深度信念网络对高维特征降维并实现运动想象模式的识别, 避免了人工确定想象最优时段及最优频段造成的信息遗漏。实验结果表明,利用变分模态分解与深度信念网络自动提取最优时段及最优频段特征的方法有效提升了运动想象识别率。

	服务

	把本文推荐给朋友
	加入我的书架
	加入引用管理器
	E-mail Alert
	RSS
	作者相关文章
	何群
	杜硕
	王煜文
	陈晓玲
	谢平

关键词 ：计量学, 脑机接口, 运动想象, 变分模态分解, 高维特征, 特征融合, 深度信念网络, 脑电信号

Abstract：The traditional method of manually determining the optimal periods and frequency bands resulted in the omission of information and the reduction of the recognition rate of motor imagery (MI). Therefore, MI became a challenging issue in brain-computer interface (BCI). Aiming at this issue, variational mode decomposition (VMD) and deep belief network (DBN) were applied to the classification of MI. VMD was proposed to decompose the electroencephalograph (EEG) into multiple narrow band components, then the marginal spectrum, the instantaneous energy spectrum and the joint time-frequency features were extracted by Hilbert transform, then these features were fused. The DBN was proposed to reduce the dimensions of fused high-dimensional features to recognize the pattern of MI, which avoided the omission of information caused by choosing the optimal periods and frequency bands manually. The results showed that the recognition accuracy of MI was improved effectively by the proposed method based on VMD and DBN to automatically extract the optimal period and frequency bands .

Key words： metrology;brain-computer interface;motor imagery;variational mode decomposition;high-dimensional features;feature fusion;deep belief network EEG signals

收稿日期: 2018-05-15 发布日期: 2019-12-19

PACS:

TB973

基金资助:国家自然科学基金面上项目(61673336);河北省自然科学基金(F2015203372)

通讯作者: 谢平(1972-),女,黑龙江齐齐哈尔人,燕山大学教授,主要从事脑-机接口、智能康复方面的研究。 Email:pingx@ysu.edu.cn E-mail: hq@ysu.edu.cn

作者简介: 何群(1969-),男,黑龙江哈尔滨人,燕山大学副教授,主要从事机械故障诊断、脑-机接口方面的研究。 Email:hq@ysu.edu.cn

引用本文:

何群,杜硕,王煜文,陈晓玲,谢平. 基于变分模态分解与深度信念网络的运动想象分类识别研究[J]. 计量学报, 2020, 41(1): 90-99.
HE Qun,DU Shuo,WANG Yu-wen,CHEN Xiao-ling,XIE Ping. The Classification of EEG Induced by Motor Imagery Based on Variational Mode Decomposition and Deep Belief Network. Acta Metrologica Sinica, 2020, 41(1): 90-99.

链接本文:

http://jlxb.china-csm.org:81/Jwk_jlxb/CN/10.3969/j.issn.1000-1158.2020.01.17 或 http://jlxb.china-csm.org:81/Jwk_jlxb/CN/Y2020/V41/I1/90

［1］Millán J D R, Rupp R, Müller-Putz G R, et al. Combining Brain-Computer Interfaces and Assistive Technologies: State-of-the-Art and Challenges［J］. Frontiers in Neuroscience, 2010, 4(1): 1-15.
［2］于向洋, 罗志增. 基于小波系数非线性连续函数衰减的脑电信号去噪［J］. 计量学报, 2017,38(6):754-757.
Yu X Y, Luo Z Z. EEG signal denoising based on a wavelet nonlinear continuous function［J］. Acta Metrologica Sinica, 2017,38(6):754-757.
［3］付荣荣,鲍甜恬,田永胜, 等. 基于子成分分解的脑电信号去噪方法比较研究［J］. 计量学报, 2019, 40(4):708-713.
Fu R R, Bao T T, Tian Y S, et al. Comparative Study on Denoising Methods of EEG Signals Based on
Subcomponent Decomposition［J］. Acta Metrologica Sinica, 2019, 40(4):708-713.
［4］Pfurtscheller G. Functional brain imaging based on ERD/ERS［J］. Vision research, 2001, 41(10-11): 1257-1260.
［5］Hsu W Y. Assembling A Multi-Feature EEG Classifier for Left-Right Motor Imagery Data Using Wavelet-Based Fuzzy Approximate Entropy for Improved Accuracy ［J］. International Journal of Neural Systems, 2015, 25(8): 1-13.
［6］张立国, 张玉曼, 金梅, 等. 基于盲源分离的运动想象脑电信号特征提取方法的研究［J］. 计量学报, 2015, 36(5): 535-539.
Zhang L G, Zhang Y M, Jin M, et al. A Research on the Method of Motor Imagery EEG Feature Extraction Based on Blind Source Separation［J］. Acta Metrologica Sinica, 2015, 36(5): 535-539.
［7］Osuagwu B A, Vuckovic A. Similarities between explicit and implicit motor imagery in mental rotation of hands: An EEG study ［J］. Neuropsychologia, 2014, 65: 197-210.
［8］王登, 苗夺谦, 王睿智. 一种新的基于小波包分解的EEG特征抽取与识别方法研究［J］. 电子学报, 2013, 41(1): 193-198.
Wang D, Miao D Q, Wang R Z. A New Method of EEG Classification with Feature Extraction Based on Wavelet Packet Decomposition ［J］. Acta Electronica Sinica, 2013, 41(1): 193-198.
［9］胡春海, 李涛, 刘永红, 等. 基于改进差分进化算法的运动想象脑机接口频带选择［J］. 计量学报, 2018, 39(2): 276-279.
Hu C H, Li T, Liu Y H, et al. Motor Imagrey Brain Computer Interface Band Selection Based on Improved Differential Evolution Algorithm［J］. Acta Metrologica Sinica, 2018, 39(25): 276-279.
［10］孙曜, 文成林, 韦巍. 基于脑电和眼电的运动想象多尺度识别方法研究［J］. 电子学报, 2018, 46(3): 714-720.
Sun Y, Wen C L, Wei W. Research on EEG and EOG Based Multiscale Recognization Method of Motor Imagery［J］. Acta Electronica Sinica, 2018, 46(3): 714-720.
［11］Bashar S K, Bhuiyan M I H. Classification of motor imagery movements using multivariate empirical mode decomposition and short time Fourier transform based hybrid method ［J］. Engineering Science & Technology, 2016, 19(3): 1457-1464.
［12］李明爱, 田晓霞, 孙炎珺, 等. 基于改进GHSOM的运动想象脑电信号自适应识别方法［J］. 仪器仪表学报, 2015, 36(5): 1064-1071.
Li M A, Tian X X, Sun Y J, et al. Adaptive recognition method based on improved-GHSOM for motor imagery EEG［J］. Chinese Journal of Scientific Instrument, 2015, 36(5): 1064-1071.
［13］Ren Y F, Wu Y. Convolutional deep belief networks for feature extraction of EEG signal［C］// IEEE. International Joint Conference on Neural Networks(IJCNN). Beijing, China, 2014: 2850-2853.
［14］Lu N, Li T, Ren X, et al. A Deep Learning Scheme for Motor Imagery Classification based on Restricted Boltzmann Machines［J］. IEEE Transactions on Neural Systems & Rehabilitation Engineering, 2016, 25(6): 566-576.
［15］Tabar Y R, Halici U. A novel deep learning approach for classification of EEG motor imagery signals［J］. Journal of Neural Engineering, 2016, 14(1): 1-11.
［16］Dragomiretskiy K, Zosso D. Variational Mode Decomposition［J］. IEEE Transactions on Signal Processing, 2014, 62(3): 531-544.
［17］Hinton G E, Salakhutdinov R R. Reducing the Dimensionality of Data with Neural Networks［J］. Science, 2006, 313(5786): 504-507.
［18］Movahedi F, Coyle J L, Sejdic′ E. Deep belief networks for electroencephalography: A review of recent contributions and future outlooks［J］. IEEE Journal of Biomedical & Health Informatics, 2017, 22(3): 642-652.
［19］李明爱, 崔燕, 杨金福, 等. 基于HHT和CSSD的多域融合自适应脑电特征提取方法［J］. 电子学报, 2013, 41(12): 2479-2486.
Li M A, Cui Y, Yang J F, et al. An Adaptive Multi-Domain Fusion Feature Extraction with Method HHT and CSSD［J］. Acta Electronica Sinica, 2013, 41(12): 2479-2486.
［20］Smith J S. The local mean decomposition and its application to EEG perception data［J］. Journal of the Royal Society Interface, 2005, 2(5): 443-454.
［21］BCI Competition II ［EB/OL］. http: //www. bbci. de/competition/ii/, 2017-05-05.
［22］BCI Competition III ［EB/OL］. http: //www. bbci. de/competition/iii/,2017-05-05.
［23］BCI Competition IV ［EB/OL］. http: //www. bbci. de/competition/iv/, 2017-05-05.
［24］Rodríguez-Bermúdez G, García-Laencina P J, Roca-González J, et al. Efficient feature selection and linear discrimination of EEG signals［J］. Neurocomputing, 2013, 115: 161-165.
［25］杨默涵, 陈万忠, 李明阳. 基于总体经验模态分解的多类特征的运动想象脑电识别方法研究［J］. 自动化学报, 2017, 43(5): 743-752.
Yang M H, Chen W Z, Li M Y. Multiple Feature Extraction Based on Ensemble Empirical Mode Decomposition for Motor Imagery EEG Recognition Tasks［J］. Acta Automatica Sinica, 2017, 43(5): 743-752.
［26］Raza H, Cecotti H. Prasad G. A combination of transductive and inductive learning for handling non-stationarities in motor imagery classification［C］//IEEE. International Joint Conference on Neural Networks(IJCNN). Vancouver, Canada, 2016: 763-770.
［27］Sayed K, Kamel M, Alhaddad M, et al. Characterization of phase space trajectories for Brain-Computer Interface［J］. Biomedical Signal Processing & Control, 2017, 38: 55-66.