DA-CapsNet: A multi-branch capsule network based on adversarial domain adaption for cross-subject EEG emotion recognition

doi:10.1016/j.knosys.2023.111137

CORC > 自动化研究所 > 中国科学院自动化研究所 > 多模态人工智能系统全国重点实验室

	DA-CapsNet: A multi-branch capsule network based on adversarial domain adaption for cross-subject EEG emotion recognition
	Liu, Shuaiqi 1,2,3; Wang, Zeyao 1,4; An, Yanling 5; Li, Bing 3; Wang, Xinrui 1,4; Zhang, Yudong 6
刊名	KNOWLEDGE-BASED SYSTEMS
	2024-01-11
卷号	283 页码:12
关键词	EEG emotion recognition Capsule network Adversarial domain adaptation Transfer learning
ISSN号	0950-7051
DOI	10.1016/j.knosys.2023.111137
通讯作者	Liu, Shuaiqi(shdkj-1918@163.com)
英文摘要	Due to inter-individual variances, cross-subject electroencephalogram (EEG)-based emotion recognition is a challenging task. In this paper, we construct a multi-branch Capsule network (named DA-CapsNet) based on domain adaptation to improve the performance of cross-subject EEG emotion recognition. To fully capture the various intensity characteristics of a single emotion, firstly, DA-CapsNet decomposes the source and the target domain EEG signals into four frequency bands and homomorphically groups the data in each band, and then extracts the differential entropy (DE) features for each group separately. Taking into account the spatial arrangement of the electrodes, the DE features are mapped into a two-dimensional matrix to form a homomorphic difference cube sequence (HDCS). Second, to enhance the feature information of the same emotion and accelerate the run efficiency of the network, a parallel structured multi-branch primary Capsual network (CapsNet) is constructed in this paper. The multi-branch primary CapsNet can effectively extract the aforementioned sequence discriminative features and fuse them as the input features of the capsule emotion classifier. Finally, to lessen inter-domain distribution discrepancies, we brought adversarial domain adaptation to improve the performance of cross-subject emotion recognition. Numerous tests are run on the three public datasets of EEG, and the results show that the proposed algorithm in this paper works well.
资助项目	National Natural Science Foundation of China[62172139] ; Natural Science Foundation of Hebei Province[F2022201055] ; Research Project of Hebei University Intelligent Financial Application Technology R D Center[XGZJ2022022] ; Science Foundation Science Research Project of Hebei Province[DXK202102] ; Natural Science Interdisciplinary Research Program of Hebei University[IT2023B05] ; Hebei University Research and Innovation Team Support Project[202200007] ; Open Project Program of the National Laboratory of Pattern Recognition (NLPR)[HBU2022ss042] ; Post-graduate's Innovation Fund Project of Hebei University[2020GDDSIPL-04] ; Open Foundation of Guangdong Key Laboratory of Digital Signal and Image Processing Technology[2022M713361] ; High-Perfor-mance Computing Center of Hebei University ; Project Funded by China Postdoctoral ; [BJ2020030]
WOS研究方向	Computer Science
语种	英语
出版者	ELSEVIER
WOS记录号	WOS:001112136900001
资助机构	National Natural Science Foundation of China ; Natural Science Foundation of Hebei Province ; Research Project of Hebei University Intelligent Financial Application Technology R D Center ; Science Foundation Science Research Project of Hebei Province ; Natural Science Interdisciplinary Research Program of Hebei University ; Hebei University Research and Innovation Team Support Project ; Open Project Program of the National Laboratory of Pattern Recognition (NLPR) ; Post-graduate's Innovation Fund Project of Hebei University ; Open Foundation of Guangdong Key Laboratory of Digital Signal and Image Processing Technology ; High-Perfor-mance Computing Center of Hebei University ; Project Funded by China Postdoctoral
内容类型	期刊论文
源URL	[http://ir.ia.ac.cn/handle/173211/55140]
专题	多模态人工智能系统全国重点实验室
通讯作者	Liu, Shuaiqi
作者单位	1.Hebei Univ, Coll Elect & Informat Engn, Baoding 071000, Hebei, Peoples R China 2.Machine Vis Technol Innovat Ctr Hebei Prov, Baoding 071000, Peoples R China 3.Chinese Acad Sci, Inst Automat, State Key Lab Multimodal Artificial Intelligence S, Beijing 100190, Peoples R China 4.Key Lab Digital Med Engn Hebei Prov, Baoding 071002, Peoples R China 5.Beijing Jiaotong Univ, Inst Informat Sci, Beijing 100044, Peoples R China 6.Univ Leicester, Sch Comp & Math, Leicester LE1 7RH, England
推荐引用方式 GB/T 7714	Liu, Shuaiqi,Wang, Zeyao,An, Yanling,et al. DA-CapsNet: A multi-branch capsule network based on adversarial domain adaption for cross-subject EEG emotion recognition[J]. KNOWLEDGE-BASED SYSTEMS,2024,283:12.
APA	Liu, Shuaiqi,Wang, Zeyao,An, Yanling,Li, Bing,Wang, Xinrui,&Zhang, Yudong.(2024).DA-CapsNet: A multi-branch capsule network based on adversarial domain adaption for cross-subject EEG emotion recognition.KNOWLEDGE-BASED SYSTEMS,283,12.
MLA	Liu, Shuaiqi,et al."DA-CapsNet: A multi-branch capsule network based on adversarial domain adaption for cross-subject EEG emotion recognition".KNOWLEDGE-BASED SYSTEMS 283(2024):12.