现代电子技术2026，Vol.49Issue(17)：38-45,8.DOI:10.16652/j.issn.1004-373X.2026.17.007
基于多阶段融合模型的运动想象脑电信号分类方法
A multi-stage fusion model for motor imagery EEG signal classification
摘要
Abstract
To address the challenges of insufficient temporal dependency modeling and limited channel selection capability in motor imagery electroencephalography(MI-EEG)classification,this paper designs a deep neural network architecture based on multi-stage feature fusion.The proposed model integrates time-frequency features,channel attention information,and temporal dependencies across multiple stages:a one-dimensional convolutional module inspired by EEGNet extracts local time-frequency representations;a multi-layer Transformer encoder captures global temporal dependencies;an SE(Squeeze-and-Excitation)block enhances responses from key brain regions;followed by 1×1 convolutions and residual connections for spatial feature mapping and fusion;and finally,a temporal convolutional network(TCN)performs deep-level temporal modeling and decoding output.In within-subject experiments on the BCI Competition IV-2a and BCI Competition IV-2b datasets,the model achieved average classification accuracies of 82.75%and 87.82%,respectively.In comparison with established methods such as ShallowConvNet,DeepConvNet,EEGNet,MI-CAT and CTNet,the proposed model demonstrates superior accuracy in within-subject MI-EEG classification tasks,validating the effectiveness and generalization ability of multi-stage fusion mechanism in time-frequency feature modeling.关键词
脑电信号/运动想象/Transformer/TCN/通道注意力/特征融合/多阶段建模Key words
EEG/motor imagery/Transformer/TCN/channel attention/feature fusion/multi-stage modeling分类
信息技术与安全科学引用本文复制引用
张然,张安元,杨菁菁..基于多阶段融合模型的运动想象脑电信号分类方法[J].现代电子技术,2026,49(17):38-45,8.基金项目
吉林省教育厅科学技术研究项目:基于多模态迁移网络的假肢鲁棒控制技术研究(JJKH20250521KJ) （JJKH20250521KJ）
吉林省高等教育教学改革课题(JLJY202293196772) （JLJY202293196772）
