Filter Bank Convolutional Neural Network for SSVEP Classification

Harmonics in electroencephalogram (EEG) caused by visual stimulation are the main basis of classification of steady-state visual evoked potential (SSVEP). However, the correlation of various harmonics, which could improve the classification performance especially when evoked EEG components are much...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Dechun Zhao, Tian Wang, Yuanyuan Tian, Xiaoming Jiang
Formato:	article
Lenguaje:	EN
Publicado:	IEEE 2021
Materias:	Brain–computer interface convolutional neural networks electroencephalography filter bank steady-state visual evoked potential Electrical engineering. Electronics. Nuclear engineering TK1-9971
Acceso en línea:	https://doaj.org/article/1cb2dfddeeaf495492c57945e02b2f51
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:1cb2dfddeeaf495492c57945e02b2f51
record_format	dspace
spelling	oai:doaj.org-article:1cb2dfddeeaf495492c57945e02b2f512021-11-18T00:10:03ZFilter Bank Convolutional Neural Network for SSVEP Classification2169-353610.1109/ACCESS.2021.3124238https://doaj.org/article/1cb2dfddeeaf495492c57945e02b2f512021-01-01T00:00:00Zhttps://ieeexplore.ieee.org/document/9594841/https://doaj.org/toc/2169-3536Harmonics in electroencephalogram (EEG) caused by visual stimulation are the main basis of classification of steady-state visual evoked potential (SSVEP). However, the correlation of various harmonics, which could improve the classification performance especially when evoked EEG components are much weaker than spontaneous EEG components, has not been take into consideration in the design of classifier in previous studies. In this study, we proposed a filter bank convolutional neural network (FBCNN) method to optimize SSVEP classification. Three filters with passbands covering each harmonic of SSVEP signals are used to extract and separate the corresponding components, and the information from them are transformed into frequency domain. Subsequently, we introduce a novel convolutional neural network (CNN) architecture with three parallel CNN channels to extract and learn the harmonic features in passbands, and conclusions on the correlation among harmonics can finally be made by pair-add-up operations and dimension reductions to weigh the feature vectors. The proposed FBCNN is evaluated on two public datasets (Dataset1: 12-class, 10 subjects; Dataset2: 40-class, 35 subjects) to compare with other methods. The experimental results illustrate that FBCNN method improves the performance of CNN-based SSVEP classification methods and has a great potential to be applied in SSVEP-based BCI.Dechun ZhaoTian WangYuanyuan TianXiaoming JiangIEEEarticleBrain–computer interfaceconvolutional neural networkselectroencephalographyfilter banksteady-state visual evoked potentialElectrical engineering. Electronics. Nuclear engineeringTK1-9971ENIEEE Access, Vol 9, Pp 147129-147141 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	Brain–computer interface convolutional neural networks electroencephalography filter bank steady-state visual evoked potential Electrical engineering. Electronics. Nuclear engineering TK1-9971
spellingShingle	Brain–computer interface convolutional neural networks electroencephalography filter bank steady-state visual evoked potential Electrical engineering. Electronics. Nuclear engineering TK1-9971 Dechun Zhao Tian Wang Yuanyuan Tian Xiaoming Jiang Filter Bank Convolutional Neural Network for SSVEP Classification
description	Harmonics in electroencephalogram (EEG) caused by visual stimulation are the main basis of classification of steady-state visual evoked potential (SSVEP). However, the correlation of various harmonics, which could improve the classification performance especially when evoked EEG components are much weaker than spontaneous EEG components, has not been take into consideration in the design of classifier in previous studies. In this study, we proposed a filter bank convolutional neural network (FBCNN) method to optimize SSVEP classification. Three filters with passbands covering each harmonic of SSVEP signals are used to extract and separate the corresponding components, and the information from them are transformed into frequency domain. Subsequently, we introduce a novel convolutional neural network (CNN) architecture with three parallel CNN channels to extract and learn the harmonic features in passbands, and conclusions on the correlation among harmonics can finally be made by pair-add-up operations and dimension reductions to weigh the feature vectors. The proposed FBCNN is evaluated on two public datasets (Dataset1: 12-class, 10 subjects; Dataset2: 40-class, 35 subjects) to compare with other methods. The experimental results illustrate that FBCNN method improves the performance of CNN-based SSVEP classification methods and has a great potential to be applied in SSVEP-based BCI.
format	article
author	Dechun Zhao Tian Wang Yuanyuan Tian Xiaoming Jiang
author_facet	Dechun Zhao Tian Wang Yuanyuan Tian Xiaoming Jiang
author_sort	Dechun Zhao
title	Filter Bank Convolutional Neural Network for SSVEP Classification
title_short	Filter Bank Convolutional Neural Network for SSVEP Classification
title_full	Filter Bank Convolutional Neural Network for SSVEP Classification
title_fullStr	Filter Bank Convolutional Neural Network for SSVEP Classification
title_full_unstemmed	Filter Bank Convolutional Neural Network for SSVEP Classification
title_sort	filter bank convolutional neural network for ssvep classification
publisher	IEEE
publishDate	2021
url	https://doaj.org/article/1cb2dfddeeaf495492c57945e02b2f51
work_keys_str_mv	AT dechunzhao filterbankconvolutionalneuralnetworkforssvepclassification AT tianwang filterbankconvolutionalneuralnetworkforssvepclassification AT yuanyuantian filterbankconvolutionalneuralnetworkforssvepclassification AT xiaomingjiang filterbankconvolutionalneuralnetworkforssvepclassification
_version_	1718425210829406208

Filter Bank Convolutional Neural Network for SSVEP Classification

Ejemplares similares