Classification of covariance matrices using a Riemannian-based kernel for BCI applications

被引：302

作者：

Barachant, Alexandre ^{[1
,2
]}

Bonnet, Stephane ^{[1
]}

Congedo, Marco ^{[2
]}

Jutten, Christian ^{[2
]}

机构：

[1] CEA LETI, F-38054 Grenoble, France

[2] Grenoble Univ, CNRS, GIPSA Lab, Team ViBS Vis & Brain Signal Proc, F-38402 St Martin Dheres, France

来源：

NEUROCOMPUTING | 2013年 / 112卷

关键词：

Brain-computer interfaces; Covariance matrix; Kernel; Support vector machine; Riemannian geometry; EEG;

D O I：

10.1016/j.neucom.2012.12.039

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

The use of spatial covariance matrix as a feature is investigated for motor imagery EEG-based classification in brain-computer interface applications. A new kernel is derived by establishing a connection with the Riemannian geometry of symmetric positive definite matrices. Different kernels are tested, in combination with support vector machines, on a past BCI competition dataset. We demonstrate that this new approach outperforms significantly state of the art results, effectively replacing the traditional spatial filtering approach. (C) 2013 Elsevier B.V. All rights reserved.

引用

页码：172 / 178

页数：7

共 20 条

[1]

[Anonymous], 2012, 20th European Symposium on Artificial Neural Networks, Computational Intelligence and Machine Learning (ESANN 2012)

[2]

[Anonymous], IEEE WORKSH APPL COM

[3]

Barachant A., 2010, 2010 IEEE 12th International Workshop on Multimedia Signal Processing (MMSP), P472, DOI 10.1109/MMSP.2010.5662067

[4] Multiclass Brain-Computer Interface Classification by Riemannian Geometry [J].

Barachant, Alexandre ;

Bonnet, Stephane ;

Congedo, Marco ;

Jutten, Christian .

IEEE TRANSACTIONS ON BIOMEDICAL ENGINEERING, 2012, 59 (04) :920-928

[5] Riemannian Geometry Applied to BCI Classification [J].

Barachant, Alexandre ;

Bonnet, Stephane ;

Congedo, Marco ;

Jutten, Christian .

LATENT VARIABLE ANALYSIS AND SIGNAL SEPARATION, 2010, 6365 :629-+

[6]

Berger M., 2003, A Panoramic View of Riemannian Geometry

[7] Optimizing spatial filters for robust EEG single-trial analysis [J].

Blankertz, Benjamin ;

Tomioka, Ryota ;

Lemm, Steven ;

Kawanabe, Motoaki ;

Mueller, Klaus-Robert .

IEEE SIGNAL PROCESSING MAGAZINE, 2008, 25 (01) :41-56

[8]

Canu S., 2005, Perception Systmes et Information

[9]

Er-Yu Wang, 2010, Proceedings 7th International Symposium on Chinese Spoken Language Processing (ISCSLP 2010), P166, DOI 10.1109/ISCSLP.2010.5684490

[10] A linear feature space for simultaneous learning of spatio-spectral filters in BCI [J].

Farquhar, J. .

NEURAL NETWORKS, 2009, 22 (09) :1278-1285

← 1 2 →