基于动态脑网络特征的情绪识别方法

doi:10.19678/j.issn.1000-3428.0069633

计算机工程 ›› 2026, Vol. 52 ›› Issue (2): 125-135. doi: 10.19678/j.issn.1000-3428.0069633

基于动态脑网络特征的情绪识别方法

王海玲¹^,², 姜廷威², 方志军¹^,²^,*(), 高宇飞³

1. 江西财经大学信息管理与数学学院, 江西南昌 330013
2. 上海工程技术大学电子电气工程学院, 上海 201200
3. 郑州大学网络空间安全学院, 河南郑州 450002

收稿日期:2024-03-21 修回日期:2024-08-27 出版日期:2026-02-15 发布日期:2024-11-22
通讯作者: 方志军
作者简介:
王海玲(CCF专业会员)，女，讲师、博士，主研方向为脑电信号情感计算、深度学习
姜廷威，硕士研究生
方志军(通信作者)，教授、博士
高宇飞，副研究员、博士
基金资助:
国家自然科学基金青年基金项目(62001284); 国家自然科学基金青年基金项目(62006210); 河南省重大科技专项(221100210100); 郑州大学高层次人才科研启动项目(32340306)

Emotion Recognition Method Based on Dynamic Brain Network Feature

WANG Hailing¹^,², JIANG Tingwei², FANG Zhijun¹^,²^,*(), GAO Yufei³

1. School of Information Management and Mathematics, Jiangxi University of Finance and Economics, Nanchang 330013, Jiangxi, China
2. School of Electronic and Electrical Engineering, Shanghai University of Engineering Technology, Shanghai 201200, China
3. Academy of Cyberspace Security, Zhengzhou University, Zhengzhou 450002, Henan, China

Received:2024-03-21 Revised:2024-08-27 Online:2026-02-15 Published:2024-11-22
Contact: FANG Zhijun

摘要/Abstract

摘要：

情绪识别是人机交互(HCI)与情感智能领域的重要前沿课题之一。然而, 目前基于脑电(EGG)信号的情绪识别方法主要提取静态特征, 无法挖掘情绪的动态变化特性, 难以提升情绪识别能力。在基于EGG构建动态脑功能网络的研究中, 常采用滑动窗口方法, 通过依次构建不同窗口内的功能连接网络以形成动态网络。但该方法存在主观设定窗长的问题, 无法提取每个时间点情绪状态的连接模式, 导致时间信息丢失和脑连接信息不完整。针对上述问题, 提出动态线性相位测量(dyPLM)方法, 该方法无需使用滑窗, 即可自适应地在每个时间点构建情绪相关脑网络, 更精准地刻画情绪的动态变化特性。此外, 还提出一种卷积门控神经网络(CNGRU)情绪识别模型, 该模型可进一步提取动态脑网络深层次特征, 有效提高情绪识别准确性。在公开情绪识别脑电数据集DEAP(Database for Emotion Analysis using Physiological signals)上进行验证, 所提方法四分类准确率高达99.71%, 较MFBPST-3D-DRLF提高3.51百分点。在SEED(SJTU Emotion EEG Dataset)数据集上进行验证, 所提方法三分类准确率达到99.99%, 较MFBPST-3D-DRLF提高3.32百分点。实验结果证明了所提出的动态脑网络构建方法dyPLM和情绪识别模型CNGRU的有效性和实用性。

关键词: 脑电信号, 情绪识别, 动态脑网络, 卷积神经网络, 门控循环单元

Abstract:

Emotion recognition is one of the most important frontier research topics in the field of Human-Computer Interaction (HCI) emotional intelligence. However, at present, Electroencephalogram (EEG)-based emotion recognition extracts static features and cannot mine the dynamic characteristics of emotions, and it is difficult to improve the emotion recognition ability. In current research on the construction of dynamic Brain Functional Networks (dBFNs) using EEG signals, a sliding window is usually used to form a dBFN by sequentially constructing a functional connectivity network in different windows. However, this method is limited by subjectively setting the window length and cannot extract the connection mode of the emotional state at each time point. Therefore, the loss of temporal information leads to the loss of brain connection information. To solve these problems, this study proposes a dynamic Phase Linearity Measurement (dyPLM) method that can adaptively construct an emotion-related brain network at each time point without sliding windows to characterize the dynamic characteristics of emotions. In addition, a Convolutional Neural Gate Recurrent Unit (CNGRU) emotion recognition model is proposed that could further extract the deep-seated features of the dynamic brain network and effectively improve the accuracy of emotion recognition. In experiments on the public emotion recognition EEG dataset, Database for Emotion Analysis using Physiological signals (DEAP), the four-class classification accuracy is as high as 99.71%, and the recognition accuracy is improved by 3.51 percentage points compared to MFBPST-3D-DRLF. On the SJTU Emotion EEG Dataset (SEED), the three-class classification accuracy is 99.99%, and the recognition accuracy is improved by 3.32 percentage points compared to MFBPST-3D-DRLF. This study demonstrates the effectiveness and practicality of the proposed methods.

Key words: Electroencephalogram (EEG) signal, emotion recognition, dynamic brain network, convolutional neural network, Gate Recurrent Unit (GRU)

王海玲, 姜廷威, 方志军, 高宇飞. 基于动态脑网络特征的情绪识别方法[J]. 计算机工程, 2026, 52(2): 125-135.

WANG Hailing, JIANG Tingwei, FANG Zhijun, GAO Yufei. Emotion Recognition Method Based on Dynamic Brain Network Feature[J]. Computer Engineering, 2026, 52(2): 125-135.

https://www.ecice06.com/CN/Y2026/V52/I2/125

图/表 10

图1 CNGRU模型结构

Fig.1 The structure of CNGRU model

图2 Lang情绪二维模型

Fig.2 The two-dimensional Lang model of emotion

图3 所有被试不同特征在随机森林分类器下的识别结果平均值

Fig.3 The average of the emotion recognition results of different features of all subjects under the random forest classifier

图4 所有被试的dyPLM特征在不同分类器下的平均分类结果

Fig.4 The average classification results of dyPLM features of all subjects under different classifiers

图5 所有被试使用dyPLM特征和CNGRU模型进行情绪识别的混淆矩阵

Fig.5 The confusion matrix of all subjects using dyPLM features and CNGRU model for emotion recognition

图6 1个被试的脑网络浅层特征和深层特征

Fig.6 The superficial features and deep features of the brain network of one of the subjects

图7 不同特征在随机森林下的识别结果

Fig.7 The emotion recognition results of different features under random forest classifier

图8 所有被试在CNGRU分类器下的识别准确率

Fig.8 The emotion recognition accuracy of all subjects under the CNGRU model

参考文献 42

1	ALARCAO S M, FONSECA M J. Emotions recognition using EEG signals: a survey. IEEE Transactions on Affective Computing, 2019, 10(3): 374- 393. doi: 10.1109/TAFFC.2017.2714671
2	LI J Y, HUA H Q, XU Z H, et al. Cross-subject EEG emotion recognition combined with connectivity features and meta-transfer learning. Computers in Biology and Medicine, 2022, 145, 105519. doi: 10.1016/j.compbiomed.2022.105519
3	LI C, HOU Y M, SONG R C, et al. Multi-channel EEG-based emotion recognition in the presence of noisy labels. Science China Information Sciences, 2022, 65(4): 140405. doi: 10.1007/s11432-021-3439-2
4	王夙喆, 张雪英, 陈晓玉, 等. 基于有效注意力和GAN结合的脑卒中EEG增强算法. 计算机工程, 2024, 50(8): 336- 344. doi: 10.19678/j.issn.1000-3428.0068558
	WANG S Z, ZHANG X Y, CHEN X Y, et al. EEG enhancement algorithm based on combination of effective attention and GAN. Computer Engineering, 2024, 50(8): 336- 344. doi: 10.19678/j.issn.1000-3428.0068558
5	ALHAGRY S, FAHMY A A, EL-KHORIBI R A. Emotion recognition based on EEG using LSTM recurrent neural network. International Journal of Advanced Computer Science and Applications, 2017, 8(10): 355- 358.
6	NAJI M, FIROOZABADI M, AZADFALLAH P. Classification of music-induced emotions based on information fusion of forehead biosignals and electrocardiogram. Cognitive Computation, 2014, 6(2): 241- 252. doi: 10.1007/s12559-013-9239-7
7	PETRANTONAKIS P C, HADJILEONTIADIS L J. Emotion recognition from EEG using higher order crossings. IEEE Transactions on Information Technology in Biomedicine, 2010, 14(2): 186- 197. doi: 10.1109/TITB.2009.2034649
8	LI P, LIU H, SI Y, et al. EEG based emotion recognition by combining functional connectivity network and local activations. IEEE Transactions on Biomedical Engineering, 2019, 66(10): 2869- 2881. doi: 10.1109/TBME.2019.2897651
9	LI X, ZHANG Y, TIWARI P, et al. EEG based emotion recognition: a tutorial and review. ACM Computing Surveys, 2023, 55(4): 1- 57.
10	ZHENG W L, ZHU J Y, LU B L. Identifying stable patterns over time for emotion recognition from EEG. IEEE Transactions on Affective Computing, 2019, 10(3): 417- 429. doi: 10.1109/TAFFC.2017.2712143
11	ZHANG Y C, YAN G H, CHANG W W, et al. EEG-based multi-frequency band functional connectivity analysis and the application of spatio-temporal features in emotion recognition. Biomedical Signal Processing and Control, 2023, 79, 104157. doi: 10.1016/j.bspc.2022.104157
12	WU X, ZHENG W L, LI Z Y, et al. Investigating EEG-based functional connectivity patterns for multimodal emotion recognition. Journal of Neural Engineering, 2022, 19(1): 016012. doi: 10.1088/1741-2552/ac49a7
13	STAM C J, NOLTE G, DAFFERTSHOFER A. Phase lag index: assessment of functional connectivity from multi channel EEG and MEG with diminished bias from common sources. Human Brain Mapping, 2007, 28(11): 1178- 1193. doi: 10.1002/hbm.20346
14	BASELICE F, SORRISO A, RUCCO R, et al. Phase linearity measurement: a novel index for brain functional connectivity. IEEE Transactions on Medical Imaging, 2019, 38(4): 873- 882. doi: 10.1109/TMI.2018.2873423
15	WU X, ZHENG W L, LU B L. Identifying functional brain connectivity patterns for EEG-based emotion recognition[C]//Proceedings of the 9th International IEEE/EMBS Conference on Neural Engineering (NER). San Francisco, USA: IEEE Press, 2019: 235-238.
16	GUPTA R, FALK T H. Relevance vector classifier decision fusion and EEG graph-theoretic features for automatic affective state characterization. Neurocomputing, 2016, 174(1): 875- 884.
17	CHEN C, LI Z, WAN F, et al. Fusing frequency-domain features and brain connectivity features for cross-subject emotion recognition. IEEE Transactions on Instrumentation and Measurement, 2022, 71, 1- 15.
18	GUO W H, LI Y X, LIU M X, et al. Functional connectivity-enhanced feature-grouped attention network for cross-subject EEG emotion recognition. Knowledge-Based Systems, 2024, 283, 111199. doi: 10.1016/j.knosys.2023.111199
19	LIU X C, LI T, TANG C, et al. Emotion recognition and dynamic functional connectivity analysis based on EEG. IEEE Access, 2019, 7, 143293- 143302. doi: 10.1109/ACCESS.2019.2945059
20	O'NEILL G C, TEWARIE P K, COLCLOUGH G L, et al. Measurement of dynamic task related functional networks using MEG. NeuroImage, 2017, 146, 667- 678. doi: 10.1016/j.neuroimage.2016.08.061
21	RIZKALLAH J, ANNEN J, MODOLO J, et al. Decreased integration of EEG source-space networks in disorders of consciousness. NeuroImage: Clinical, 2019, 23, 101841. doi: 10.1016/j.nicl.2019.101841
22	NOBUKAWA S, KIKUCHI M, TAKAHASHI T. Changes in functional connectivity dynamics with aging: a dynamical phase synchronization approach. Neuroimage, 2019, 188, 357- 368. doi: 10.1016/j.neuroimage.2018.12.008
23	YANG J, CHEN S S, HUANGFU H R, et al. Dynamic functional connectivity of electroencephalogram in the resting state. Acta Physica Sinica, 2015, 64(5): 058701. doi: 10.7498/aps.64.058701
24	XEFTERIS V R, TSANOUSA A, GEORGAKOPOULOU N, et al. Graph theoretical analysis of EEG functional connectivity patterns and fusion with physiological signals for emotion recognition. Sensors, 2022, 22(21): 8198. doi: 10.3390/s22218198
25	李存波, 杨蕾, 陈昭瑾, 等. 基于脑电网络图特征的情绪识别研究. 数据采集与处理, 2023, 38(4): 815- 823.
	LI C B, YANG L, CHEN Z J, et al. Emotion recognition based on graph features extracted from EEG networks. Journal of Data Acquisition and Processing, 2023, 38(4): 815- 823.
26	STOCKWELL R G, MANSINHA L, LOWE R P. Localization of the complex spectrum: the S transform. IEEE Transactions on Signal Processing, 1996, 44(4): 998- 1001. doi: 10.1109/78.492555
27	KIRANYAZ S, INCE T, HAMILA R, et al. Convolutional neural networks for patient-specific ECG classification[C]//Proceedings of the 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society (EMBC). Washington D. C., USA: IEEE Press, 2015: 2608-2611.
28	KIRANYAZ S, AVCI O, ABDELJABER O, et al. 1D convolutional neural networks and applications: a survey. Mechanical Systems and Signal Processing, 2021, 151, 107398. doi: 10.1016/j.ymssp.2020.107398
29	LU L M, ZHANG C L, CAO K, et al. A multichannel CNN-GRU model for human activity recognition. IEEE Access, 2022, 10, 66797- 66810. doi: 10.1109/ACCESS.2022.3185112
30	ZHONG M Y, YANG Q Y, LIU Y, et al. EEG emotion recognition based on TQWT-features and hybrid convolutional recurrent neural network. Biomedical Signal Processing and Control, 2023, 79, 104211. doi: 10.1016/j.bspc.2022.104211
31	BAGHERZADEH S, SHAHABI M S, SHALBAF A. Detection of schizophrenia using hybrid of deep learning and brain effective connectivity image from electroencephalogram signal. Computers in Biology and Medicine, 2022, 146, 105570. doi: 10.1016/j.compbiomed.2022.105570
32	DEY R, SALEM F M. Gate-variants of Gated Recurrent Unit (GRU) neural networks[C]//Proceedings of the IEEE 60th International Midwest Symposium on Circuits and Systems (MWSCAS). Boston, USA: IEEE Press, 2017: 1597-1600.
33	CHUNG J, GULCEHRE C, CHO K, et al. Empirical evaluation of gated recurrent neural networks on sequence modeling[EB/OL]. [2024-01-02]. https://arxiv.org/abs/1412.3555.
34	KOELSTRA S, MUHL C, SOLEYMANI M, et al. DEAP: a database for emotion analysis; using physiological signals. IEEE Transactions on Affective Computing, 2012, 3(1): 18- 31. doi: 10.1109/T-AFFC.2011.15
35	MERT A, AKAN A. Emotion recognition from EEG signals by using multivariate empirical mode decomposition. Pattern Analysis and Applications, 2018, 21(1): 81- 89. doi: 10.1007/s10044-016-0567-6
36	SUTSKEVER I, MARTENS J, DAHL G, et al. On the importance of initialization and momentum in deep learning[C]//Proceedings of the 30th International Conference on Machine Learning. Atlanta, USA: [s. n. ], 2013: 1-10.
37	YIN Y Q, ZHENG X W, HU B, et al. EEG emotion recognition using fusion model of graph convolutional neural networks and LSTM. Applied Soft Computing, 2021, 100, 106954. doi: 10.1016/j.asoc.2020.106954
38	廉小亲, 罗志宏, 蔡沫豪, 等. 基于卷积神经网络的脑电情绪识别方法. 计算机仿真, 2022, 39(8): 268- 274.
	LIAN X Q, LUO Z H, CAI M H, et al. EEG emotion recognition method based on convolutional neural network. Computer Simulation, 2022, 39(8): 268- 274.
39	QIN C, LANLAN C, RUNQIANG J. Emotion recognition of EEG based on ensemble CapsNet. Journal of Computer Engineering & Applications, 2022, 58(8): 175- 184.
40	ZHENG W, PAN B. A spatiotemporal symmetrical transformer structure for EEG emotion recognition. Biomedical Signal Processing and Control, 2024, 87, 105487. doi: 10.1016/j.bspc.2023.105487
41	MIAO M M, ZHENG L X, XU B G, et al. A multiple frequency bands parallel spatial—temporal 3D deep residual learning framework for EEG-based emotion recognition. Biomedical Signal Processing and Control, 2023, 79, 104141. doi: 10.1016/j.bspc.2022.104141
42	XIAO G W, SHI M, YE M W, et al. 4D attention-based neural network for EEG emotion recognition. Cognitive Neurodynamics, 2022, 16(4): 805- 818. doi: 10.1007/s11571-021-09751-5

[1]	陈振清, 万加富, 张锐. 融合时序依赖性与数据特征的自适应无损分段压缩方法[J]. 计算机工程, 2026, 52(2): 79-88.
[2]	王庆荣, 郝福乐, 朱昌锋, 王俊杰. 基于多特征融合的车辆轨迹预测研究[J]. 计算机工程, 2026, 52(2): 331-341.
[3]	武东辉, 王金凤, 仇森, 刘国志. 基于EWBiLSTM-ATT的数据手套手语识别[J]. 计算机工程, 2025, 51(8): 107-119.
[4]	武东辉, 王金凤, 仇森, 刘国志. 基于EWBiLSTM-ATT的数据手套手语识别[J]. 计算机工程, 2025, 51(8): 107-119.
[5]	田银花, 杨立飞, 韩咚, 杜玉越. 基于改进BERT和轻量化CNN的业务流程合规性检查方法[J]. 计算机工程, 2025, 51(7): 199-209.
[6]	柳大格, 游进国, 耿齐祁. 融合全局与局部语义的跨领域方面词抽取[J]. 计算机工程, 2025, 51(6): 116-126.
[7]	杨萍, 张汐. 改进DeepLabv3+的道路表面裂缝检测方法[J]. 计算机工程, 2025, 51(4): 261-270.
[8]	张肇鑫, 黄世泽, 张兵杰, 沈拓. 面向交通场景的运动模糊伪装对抗样本生成方法[J]. 计算机工程, 2025, 51(3): 45-53.
[9]	胡书林, 张华军, 邓小涛, 王征华. 结合依存图卷积的中文文本相似度计算研究[J]. 计算机工程, 2025, 51(3): 76-85.
[10]	张树华, 王继业, 赵传奇, 陈宏铭, 郭咏雯. 面向输电线路边缘智能的硬件加速设计[J]. 计算机工程, 2025, 51(2): 213-222.
[11]	贾硕, 林士飏, 杨苗会, 孙滕. 改进鲸鱼优化GRU的窄路短时车流量预测[J]. 计算机工程, 2025, 51(2): 111-125.
[12]	徐磊, 曾艳, 袁俊峰, 岳鲁鹏, 殷昱煜, 张纪林, 薛梅婷, 韩猛. 基于自注意力机制的时间序列插补[J]. 计算机工程, 2025, 51(11): 90-99.
[13]	郑洁云, 张章煌, 宣菊琴, 魏鑫, 薛静玮. 基于知识图谱和图卷积神经网络的配电网智能规划方法[J]. 计算机工程, 2025, 51(11): 392-402.
[14]	周嘉文, 郑小盈, 祝永新, 林思敏, 陈凌曜, 曾洪斌, 郭俞, 王馨莹. 多头自注意力与双线性池化融合的心肌缺血影像分类[J]. 计算机工程, 2025, 51(11): 246-257.
[15]	肖志鹏, 何书峰, 田春岐. EmoRepLKNet: 一种基于UniRepLKNet的面部情绪识别神经网络[J]. 计算机工程, 2025, 51(11): 54-62.

选择文件类型/文献管理软件名称

选择包含的内容

基于动态脑网络特征的情绪识别方法

Emotion Recognition Method Based on Dynamic Brain Network Feature

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 10

参考文献 42

相关文章 15

编辑推荐

Metrics

本文评价

模态框（Modal）标题

选择文件类型/文献管理软件名称

选择包含的内容

基于动态脑网络特征的情绪识别方法

Emotion Recognition Method Based on Dynamic Brain Network Feature

RichHTML

PDF

可视化

被引次数

摘要/Abstract

引用本文

使用本文

图/表 10

参考文献 42

相关文章 15

编辑推荐

Metrics

本文评价