基于重构迁移子空间多视角领域适应的脑电情感分类方法

韩少勇; 周国华; 殷新春

doi:10.12263/DZXB.20250486

您当前的位置：

首页 >

文章列表页 >

基于重构迁移子空间多视角领域适应的脑电情感分类方法

学术论文 | 更新时间：2025-12-27

- 基于重构迁移子空间多视角领域适应的脑电情感分类方法
- Reconstructed Transfer Subspace Based Multi-View Domain Adaptation Method for Electroencephalogram Emotion Classification
- 电子学报 2025年53卷第8期页码：2830-2842
- 作者机构：
  
  1.铜陵学院数学与计算机学院，安徽铜陵 244061
  2.郑州银行博士后科研工作站，河南郑州 450015
  3.常州工业职业技术学院信息工程学院，江苏常州 213164
  4.扬州大学信息工程学院，江苏扬州 225127
- 作者简介：
  
  [ "韩少勇男，1982年10月生，甘肃武威人，博士.现为铜陵学院数学与计算机学院副教授.主要研究方向为智能信息处理研究. E-mail: hsytlu@126.com" ]
  [ "周国华男，1977年7月生，江苏东台人.现为常州工业职业技术学院信息工程学院副教授.主要研究方向为智能学习、模式识别研究. E-mail: tiddyddd@sina.com.cn" ]
  [ "殷新春男，1962年2月生，江苏姜堰人，博士.现为扬州大学信息工程学院教授，博士生导师.主要研究方向为信息安全、软件质量保障、高性能计算. E-mail: xcyin@yzu.edu.cn" ]
- 基金信息：
  
  国家自然科学基金(62032005);铜陵学院人才资助项目(2024tlxyrc020);铜陵学院科研项目(2025tlxyxdz051)
- DOI：10.12263/DZXB.20250486
  中图分类号： TP391.41;
- 收稿：2025-06-06，
  
  录用：2025-07-22，
  
  纸质出版：2025-08-25
- 稿件说明：
移动端阅览
韩少勇, 周国华, 殷新春. 基于重构迁移子空间多视角领域适应的脑电情感分类方法[J]. 电子学报, 2025, 53(08): 2830-2842.

HAN Shao-yong, ZHOU Guo-hua, YIN Xin-chun. Reconstructed Transfer Subspace Based Multi-View Domain Adaptation Method for Electroencephalogram Emotion Classification[J]. Acta Electronica Sinica, 2025, 53(08): 2830-2842.
韩少勇, 周国华, 殷新春. 基于重构迁移子空间多视角领域适应的脑电情感分类方法[J]. 电子学报, 2025, 53(08): 2830-2842. DOI：10.12263/DZXB.20250486

HAN Shao-yong, ZHOU Guo-hua, YIN Xin-chun. Reconstructed Transfer Subspace Based Multi-View Domain Adaptation Method for Electroencephalogram Emotion Classification[J]. Acta Electronica Sinica, 2025, 53(08): 2830-2842. DOI：10.12263/DZXB.20250486

摘要

情感识别是人机交互智能化的关键环节.脑电（ElectroEncephaloGram，EEG）信号因其蕴含丰富的生物信息且难以伪装，成为情感分析的重要载体.然而，EEG信号特征复杂多变，且存在显著的个体间差异和时变性，导致传统机器学习方法的情感分类准确率低、泛化能力差.针对这一挑战，本文提出了一种基于重

构迁移子空间多视角领域适应（Reconstructed Transfer Subspace based Multi-View Domain Adaptation，RTS-MVDA）方法.该方法将不同特征视为独立视角，通过多视角学习探索各视角的独特性和重要性，并探索其互补关系.其核心在于将源域与目标域的多视角数据投影到一个带有低秩约束的重构迁移子空间.在该子空间中，RTS-MVDA一方面利用重构项恢复原始数据信息，并通过低秩表示保留主要判别信息；另一方面，RTS-MVDA实施线性变换对齐源域和目标域，减少领域间的分布差异.此外，RTS-MVDA构建多视角监督判别项和全局结构保持项，多视角监督判别项利用源域标签信息增强类内紧凑性和类间分离性，全局结构保持项保持数据在迁移子空间中的全局结构分布，从而更有效地将源域的判别知识迁移至目标域.在公开DEAP（Database for Emotion Analysis using Physiological signals）数据集上的实验验证表明：所提RTS-MVDA方法在唤醒度和效价维度上分别达到了73.15%和72.91%的平均准确率，其Precision、Recall和

-score指标均显著优于相关对比方法，有效提升了跨被试EEG情感识别的准确性和泛化能力.

Abstract

Emotion recognition is the key link of intelligent human-computer interaction. Electroencephalogram (EEG) has become an important carrier of emotion analysis because it contains rich biological information and is difficult to disguise. However

EEG signal features are complex and changeable

and there are significant individual differences and time variability

which lead to low accuracy and poor generalization ability of traditional machine learning methods. To address these challenges

this paper proposes a reconstructed transfer subspace based multi view domain adaptation (RTS-MVDA). This method regards different features as independent perspectives

explores the uniqueness and importance of each perspective through multi perspective learning

and mining their complementary relationship. Its core is to project the multi view data of the source domain and the target domain into a reconstruction migration subspace with low-rank constraints. In this subspace

RTS-MVDA

on the one hand

uses the reconstructed items to restore the original data information

and retains the main discrimination information through the low-rank representation; on the other hand

RTS-MVDA implements linear transformation to align the source domain and target domain

reducing the distribution difference between domains. In addition

RTS-MVDA constructs multi view supervised discriminant

and global structure preserving item. The former uses source domain label information to enhance intra class compactness and inter class separation

while the latter maintains the global structure distribution of data in the migration subspace

so as to more effectively migrate the discriminant knowledge of the source domain to the target domain. The experimental verification on the public database for emotion analysis using physiological signals (DEAP) dataset shows that the average accuracy of the proposed RTS-MVDA method in arousal and valence is 73.15% and 72.91%

respectively. Its precision

recall and

-score are significantly better than the related comparison methods

effectively improving the accuracy and generalization ability of cross-subject EEG emotion recognition.

关键词

Keywords

references

揭丽琳 , 刘勇 , 王铭勋 , 等 . 基于改进TCNN算法的脑电动态连续情绪识别研究 [J ] . 电子学报 , 2025 , 53 ( 4 ): 1347 - 1360 .

JIE L L , LIU Y , WANG M X , et al . Research on dynamic continuous emotional recognition of EEG based on improved TCNN algorithm [J ] . Acta Electronica Sinica , 2025 , 53 ( 4 ): 1347 - 1360 . (in Chinese)

师硕 , 覃嘉俊 , 于洋 , 等 . 基于改进ConvMixer和动态焦点损失的视听情感识别 [J ] . 电子学报 , 2024 , 52 ( 8 ): 2824 - 2835 .

SHI S , QIN J J , YU Y , et al . Improved ConvMixer and focal loss with dynamic weight for audio-visual emotion recognition [J ] . Acta Electronica Sinica , 2024 , 52 ( 8 ): 2824 - 2835 . (in Chinese)

张晶 , 王翌歆 , 任永功 . 统一全局空间表达的脑电信号跨被试情感识别 [J ] . 电子学报 , 2023 , 51 ( 5 ): 1396 - 1404 .

ZHANG J , WANG Y X , REN Y G . Unified global spatial representation for EEG subject-independent emotion recognition [J ] . Acta Electronica Sinica , 2023 , 51 ( 5 ): 1396 - 1404 . (in Chinese)

HAMZAH H ALI , ABDALLA K K . EEG-based emotion recognition systems; Comprehensive study [J ] . Heliyon , 2024 , 10 ( 10 ): e31485 .

ROSHANAEI M , NOROUZI H , ONTON J , et al . EEG-based functional and effective connectivity patterns during emotional episodes using graph theoretical analysis [J ] . Scientific Reports , 2025 , 15 : 2174 .

GARG D , VERMA G K , SINGH A K . EEG-based emotion recognition using mobilenet recurrent neural network with time-frequency features [J ] . Applied Soft Computing , 2024 , 154 : 111338 .

李明爱 , 张圆圆 . 基于连续小波变换和符号传递熵的脑功能网络构建方法 [J ] . 电子学报 , 2022 , 50 ( 7 ): 1600 - 1608 .

LI M A , ZHANG Y Y . A brain functional network based on continuous wavelet transform and symbolic transfer entropy [J ] . Acta Electronica Sinica , 2022 , 50 ( 7 ): 1600 - 1608 . (in Chinese)

席旭刚 , 王成浩 , 汪婷 , 等 . 基于表面肌电分解的皮层肌肉耦合机理研究 [J ] . 电子学报 , 2024 , 52 ( 8 ): 2607 - 2617 .

XI X G , WANG C H , WANG T , et al . Mechanism of corticomuscular coupling based on surface electromyography decomposition [J ] . Acta Electronica Sinica , 2024 , 52 ( 8 ): 2607 - 2617 . (in Chinese)

SAMAL P , HASHMI M F . Role of machine learning and deep learning techniques in EEG-based BCI emotion recognition system: A review [J ] . Artificial Intelligence Review , 2024 , 57 ( 3 ): 50 .

CHEN C Q , LI Z C , WAN F , et al . Fusing frequency-domain features and brain connectivity features for cross-subject emotion recognition [J ] . IEEE Transactions on Instrumentation and Measurement , 2022 , 71 : 2508215 .

SOWMYA M , VARMA P , KATARAPU D . Feature extraction from EEG signals of the brain using spectral entropy feature extraction technique [J ] . Grenze International Journal of Engineering & Technology , 2024 , 15 ( 6 ): 10 .

LI C , BIAN N , ZHAO Z P , et al . Multi-view domain-adaptive representation learning for EEG-based emotion recognition [J ] . Information Fusion , 2024 , 104 : 102156 .

XIAO Y , ZHANG Y M , PENG X P , et al . Multi-source EEG emotion recognition via dynamic contrastive domain adaptation [J ] . Biomedical Signal Processing and Control , 2025 , 102 : 107337 .

AI Q S , WANG C H , CHEN K , et al . Multi-source domain separation adversarial domain adaptation for EEG emotion recognition [J ] . Biomedical Signal Processing and Control , 2025 , 109 : 108016 .

SHI X S , SHE Q S , FANG F , et al . Enhancing cross-subject EEG emotion recognition through multi-source manifold metric transfer learning [J ] . Computers in Biology and Medicine , 2024 , 174 : 108445 .

IMTIAZ M N , KHAN N . Enhanced cross-dataset electroencephalogram-based emotion recognition using unsupervised domain adaptation [J ] . Computers in Biology and Medicine , 2025 , 184 : 109394 .

TAO J W , YAN L D , HE T . Domain-invariant adaptive graph regularized label propagation for EEG-based emotion recognition [J ] . IEEE Access , 2024 , 12 : 126774 - 126792 .

LIU W C , GUO C J , GAO C . A cross-session motor imagery classification method based on Riemannian geometry and deep domain adaptation [J ] . Expert Systems with Applications , 2024 , 237 : 121612 .

CHEN P Y , LIU X F , MA C , et al . Unsupervised domain adaptation with synchronized self-training for cross- domain motor imagery recognition [J ] . IEEE Journal of Biomedical and Health Informatics , 2025 , 29 ( 5 ): 3664 - 3677 .

JALILPOUR S , MÜLLER-PUTZ G . A framework for Interpretable deep learning in cross-subject detection of event-related potentials [J ] . Engineering Applications of Artificial Intelligence , 2025 , 139 : 109642 .

HE R F , HUANG F , MA J S , et al . Few-shot cross domain event discovery in narrative text [J ] . Information Processing & Management , 2025 , 62 ( 1 ): 103901 .

CHEN H T , XIE J L . EEG-based TSK fuzzy graph neural network for driver drowsiness estimation [J ] . Information Sciences , 2024 , 679 : 121101 .

LUO Y , LIU W , LI H Q , et al . A cross-scenario and cross-subject domain adaptation method for driving fatigue detection [J ] . Journal of Neural Engineering , 2024 , 21 ( 4 ): 046004 .

KUMAR S , ALAWIEH H , RACZ F S , et al . Transfer learning promotes acquisition of individual BCI skills [J ] . PNAS Nexus , 2024 , 3 ( 2 ): 76 .

DARVISHI-BAYAZI M J , GHAEMI M S , LESORT T , et al . Amplifying pathological detection in EEG signaling pathways through cross-dataset transfer learning [J ] . Computers in Biology and Medicine , 2024 , 169 : 107893 .

KOELSTRA S , MUHL C , SOLEYMANI M , et al . DEAP: A database for emotion Analysis; Using physiological signals [J ] . IEEE Transactions on Affective Computing , 2012 , 3 ( 1 ): 18 - 31 .

XU Y , FANG X Z , WU J , et al . Discriminative transfer subspace learning via low-rank and sparse representation [J ] . IEEE Transactions on Image Processing , 2016 , 25 ( 2 ): 850 - 863 .

YANG L R , ZHOU Q H . Transfer subspace learning joint low-rank representation and feature selection [J ] . Multimedia Tools and Applications , 2022 , 81 ( 27 ): 38353 - 38373 .

LI S K , SONG P , ZHENG W M . Multi-source discriminant subspace alignment for cross-domain speech emotion recognition [J ] . IEEE/ACM Transactions on Audio , Speech and Language Processing, 2023 , 31 : 2448 - 2460 .

RICE J R , GARDINER J D , LAUB A J , et al . Solution of the sylvester matrix equation AXB T + CXD T = E [J ] . ACM Transactions on Mathematical Software , 1992 , 18 ( 2 ): 223 - 231 .

YANG F , ZHAO X C , JIANG W G , et al . Multi-method fusion of cross-subject emotion recognition based on high-dimensional EEG features [J ] . Frontiers in Computational Neuroscience , 2019 , 13 : 53 .

AL-FAHOUM A S , AL-FRAIHAT A A . Methods of EEG signal features extraction using linear analysis in frequency and time-frequency domains [J ] . International Scholarly Research Notices , 2014 , 2014 ( 1 ): 730218 .

THIRUMARAI SELVI C , JAYASHEELA M , AMUDHA J , et al . An EEG-based thought recognition using pseudo-Wigner-kullback-leibler deep neural classification [J ] . Circuits, Systems, and Signal Processing , 2023 , 42 ( 2 ): 1063 - 1082 .

XIAO M H , BO S . Electroencephalogram emotion recognition via AUC maximization [J ] . Algorithms , 2024 , 17 ( 11 ): 489 .

PATEL P , R R , ANNAVARAPU R N . EEG-based human emotion recognition using entropy as a feature extraction measure [J ] . Brain Informatics , 2021 , 8 ( 1 ): 20 .

LI J W , FENG G Y , LING C , et al . A resource-efficient multi-entropy fusion method and its application for EEG-based emotion recognition [J ] . Entropy , 2025 , 27 ( 1 ): 96 .

PAPPALETTERA C , CACCIOTTI A , NUCCI L , et al . Approximate entropy analysis across electroencephalographic rhythmic frequency bands during physiological aging of human brain [J ] . GeroScience , 2023 , 45 ( 2 ): 1131 - 1145 .

HUANG C Q , CHUNG F L , WANG S T . Multi-view L2-SVM and its multi-view core vector machine [J ] . Neural Networks , 2016 , 75 : 110 - 125 .

BARAJAS MONTIEL S E , MORALES E F , ESCALANTE H J . Multi-view learning forEEG signal classification ofImagined speech [C ] // Pattern Recognition . Cham : Springer , 2022 : 191 - 200 .

ZHAN S H , SUN W J , KANG P P . Robust latent common subspace learning for transferable feature representation [J ] . Electronics , 2022 , 11 ( 5 ): 810 .

LI A D , DENG Z H , ZHANG W , et al . Multiview transfer representation learning with TSK fuzzy system for EEG epilepsy detection [J ] . IEEE Transactions on Fuzzy Systems , 2024 , 32 ( 1 ): 38 - 52 .

FU Z Z , ZHANG B N , HE X R , et al . Emotion recognition based on multi-modal physiological signals and transfer learning [J ] . Frontiers in Neuroscience , 2022 , 16 : 1000716 .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

分布式目标的子空间双门限GLRT CFAR检测

基于子空间的块稀疏信号压缩感知重构算法

基于子空间三角不等式的高维码字搜索算法

基于联合特征空间投影的SAR图像域杂波抑制