DocumentCode :
3494769
Title :
Common Spatial Pattern revisited by Riemannian geometry
Author :
Barachant, Alexandre ; Bonnet, Stphane ; Congedo, Marco ; Jutten, Christian
Author_Institution :
LETI, CEA, Grenoble, France
fYear :
2010
fDate :
4-6 Oct. 2010
Firstpage :
472
Lastpage :
476
Abstract :
This paper presents a link between the well known Common Spatial Pattern (CSP) algorithm and Riemannian geometry in the context of Brain Computer Interface (BCI). It will be shown that CSP spatial filtering and Log variance features extraction can be resumed as a computation of a Riemann distance in the space of covariances matrices. This fact yields to highlight several approximations with respect to the space topology. According to these conclusions, we propose an improvement of classical CSP method.
Keywords :
brain-computer interfaces; covariance matrices; feature extraction; spatial filters; Riemannian geometry; brain computer interface; common spatial pattern algorithm; covariance matrix; log variance feature extraction; spatial filtering; Approximation methods; Covariance matrix; Eigenvalues and eigenfunctions; Electroencephalography; Feature extraction; Geometry; Symmetric matrices;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Multimedia Signal Processing (MMSP), 2010 IEEE International Workshop on
Conference_Location :
Saint Malo
Print_ISBN :
978-1-4244-8110-1
Electronic_ISBN :
978-1-4244-8111-8
Type :
conf
DOI :
10.1109/MMSP.2010.5662067
Filename :
5662067
Link To Document :
بازگشت