Abstract
A General Common Spatial Patterns for EEG Analysis With Applications to Vigilance Detection
Highlights
Electroencephalography(EEG) is the physiological method of choice to record the electrical activity generated by the brain via electrodes placed on the scalp surface
We propose a novel multiclass Common spatial pattern (CSP) (MCSP) that has an explantation of minimizing Bayesian classification error, but with a different optimization criteria from that in [28], and extend this multi-class CSP to nonparametric multi-class CSP (NMCSP)
Given samples corresponding to two classes in a high-dimensional space, CSP finds projection directions (i.e.,spatial filters) that maximize the variance for one class and at the same time minimize the variance for the other class
Summary
Electroencephalography(EEG) is the physiological method of choice to record the electrical activity generated by the brain via electrodes placed on the scalp surface. Zheng and Lin [28] presented an optimal multi-class CSP based on the bayesian error estimation for EEG feature extraction, our proposed general common spatial pattern algorithm is motivated by this method but we have different criteria function. We propose a new CSP algorithm to extract the EEG feature for vigilance estimation. We apply our proposed algorithms to the vigilance estimation as well as the motor imagery task to measure their performance on feature extraction task, comparisons are made with the multi-class CSP in [28], ACCSP in [31], RCSP in [23] as well as the CSP-L1 [34]. We validate the performance of our proposed new CSP algorithms on the designed vigilance estimation as well as the motor imagery experiments.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
