Year: 2017
Type: article
Abstract: Thispaper presents a novel method for electroencephalography (EEG) based motor imagery classification for brain computer interface (BCI) implementation using the potential features extracted bandspecific common spatial pattern (CSP).The recorded EEG signal is bandpass-filtered into multiple subbands to capture the related rhythmic components of brain signals.The CSP features are then extracted from each of these bands.The linear discriminant analysis (LDA) based classifier is subsequently used to classify the relevant subband of EEG using the features extracted by CSP.Then the highest discrimination score among all the subbands is used as the norm for overall EEG classification.The experimental results show that the proposed method yields comparatively superiorclassification performance compared to prevailing methods.
Authors: Md. Sujan Ali, Mst. Jannatul Ferdous
Institution: Jatiya Kabi Kazi Nazrul Islam University
Language: English
Subfield: Cognitive Neuroscience
Field: Neuroscience
Domain: Life Sciences
Open Access status: bronze
OpenAlex ID: W2610328114