柳长源, 李文强, 毕晓君. 基于RCNN-LSTM的脑电情感识别研究. 自动化学报, 2022, 48(3): 917−925 doi: 10.16383/j.aas.c190357 引用本文: 柳长源, 李文强, 毕晓君. 基于RCNN-LSTM的脑电情感识别研究. 自动化学报, 2022, 48 (3): 917−925 doi: 10.16383/j.aas.c190357 Liu Chang-Yuan, Li Wen-Qiang, Bi Xiao-Jun. Research on EEG emotion recognition based on RCNN-LSTM. Acta Automatica Sinica, 2022, 48(3): 917−925 doi: 10.16383/j.aas.c190357 Citation: Liu Chang-Yuan, Li Wen-Qiang, Bi Xiao-Jun. Research on EEG emotion recognition based on RCNN-LSTM. Acta Automatica Sinica, 2022, 48 (3): 917−925 doi: 10.16383/j.aas.c190357 柳长源, 李文强, 毕晓君. 基于RCNN-LSTM的脑电情感识别研究. 自动化学报, 2022, 48(3): 917−925 doi: 10.16383/j.aas.c190357 引用本文: 柳长源, 李文强, 毕晓君. 基于RCNN-LSTM的脑电情感识别研究. 自动化学报, 2022, 48 (3): 917−925 doi: 10.16383/j.aas.c190357 Liu Chang-Yuan, Li Wen-Qiang, Bi Xiao-Jun. Research on EEG emotion recognition based on RCNN-LSTM. Acta Automatica Sinica, 2022, 48(3): 917−925 doi: 10.16383/j.aas.c190357 Citation: Liu Chang-Yuan, Li Wen-Qiang, Bi Xiao-Jun. Research on EEG emotion recognition based on RCNN-LSTM. Acta Automatica Sinica, 2022, 48 (3): 917−925 doi: 10.16383/j.aas.c190357 作者简介:

柳长源：哈尔滨理工大学测控技术与通信工程学院副教授. 主要研究方向为机器学习方法研究与改进, 智能优化算法及脑电信号智能诊断技术. 本文通信作者.E-mail: [email protected]

李文强：哈尔滨理工大学测控技术与通信工程学院硕士研究生. 主要研究方向为深度学习与情感识别. E-mail: [email protected]

毕晓君：中央民族大学信息工程学院教授, 中国人工智能学会自然计算专委会成员, 黑龙江省生物医学工程学会常务副理事长. 主要研究方向为信息智能处理技术, 深度学习及智能优化算法.E-mail: [email protected]

Author Bio: LIU Chang-Yuan Associate professor at the College of Measurement and Control Technology and Communication Engineering, Harbin University of Science and Technology. His research interest covers research and improvement of machine learning methods, intelligent optimization algorithms, and EEG intelligent diagnosis technology. Corresponding author of this paper

LI Wen-Qiang Master student at the College of Measurement and Control Technology and Communication Engineering, Harbin University of Science and Technology. His research interest covers deep learning and emotion recognition

BI Xiao-Jun Professor at the School of Information Engineering, Minzu University of China, member of the Natural Computing Committee of Chinese Artificial Intelligence Society, and Executive Vice Chairman of Heilongjiang Biomedical Engineering Society. Her research interest covers information intelligent processing technology, deep learning, and intelligent optimization algorithm

情感作为人脑的高级功能, 对人们的个性特征和心理健康有很大的影响, 利用网上公开的脑电情感数据库(DEAP (Database for emotion analysis using physiological signals)数据库), 根据心理效价和激励唤醒度等级进行情感划分, 对压力和平静等5种情感进行研究分析. 针对脑电信号时空特征结合的特点, 把深度学习中的卷积神经网络(Convolutional neural network, CNN)和长短期记忆网络(Long short term memory, LSTM)两者作为基本前提, 并在此基础之上设计了一个RCNN-LSTM的脑电情感信号分类模型. 利用循环卷积神经网络(Recurrent convolutional neural network, RCNN)自动提取脑电信号中的抽象特征, 省去了人工选择与降维的过程, 然后结合LSTM网络对脑电情感信号进行分类识别. 实验结果表明, 利用该方法对5种情感类别的平均分类识别率达到了96.63%, 证明了该方法的有效性. 脑电信号 /  情感识别 /  循环卷积神经网络 /  长短期记忆神经网络 Abstract: As an advanced function of the human brain, emotion has a great influence on people＇s personality characteristics and mental health. Using the online public EEG (electroencephalogram) database (DEAP (database for emotion analysis using physiological signals) database), emotions are divided according to psychological valence and arousal level. Five emotions, including stress and calm, are studied and analyzed. According to the combination of temporal and spatial features of EEG signals, CNN (convolutional neural network) and LSTM (long short term memory) in deep learning are taken as the basic prerequisites. On the basis of this, a new and more powerful EEG signal classification model is constructed. Use the recurrent convolutional neural network (RCNN) to automatically extract the abstract features of the EEG signal, eliminate the process of manual selection and dimensional reduction, and then send it to the LSTM network to classify and identify EEG signals. The experimental results show that the average accuracy of this method is 96.63%, which demonstrates the effectiveness of this method. Key words: Electroencephalogram (EEG) /  emotion recognition /  recurrentconvolutional neural network (RCNN) /  long short term memory (LSTM)  耿雪青, 佘青山, 韩笑, 孟明. 基于人工蜂群优化高斯过程的运动想象脑电信号分类. 传感技术学报, 2017, 30(03): 378-384 doi: 10.3969/j.issn.1004-1699.2017.03.008

Geng Xue-Qing, She Qing-Shan, Han Xiao, Meng Ming. Classification of motor imagery EEG based on gaussian process optimized with artificial bee colony. Chinese Journal of Sensors and Actuators, 2017, 30(03): 378-384 doi: 10.3969/j.issn.1004-1699.2017.03.008 王薇蓉, 张雪英, 孙颖, 畅江. 关于脑电信号的情感识别优化仿真. 计算机仿真, 2018, 35(6): 426-431 doi: 10.3969/j.issn.1006-9348.2018.06.093

Wang Wei-Rong, Zhang Xue-Ying, Sun Ying, Chang Jiang. Emotion optimization identification emulation for EEG signal. Computer Simulation, 2018, 35(6): 426-431 doi: 10.3969/j.issn.1006-9348.2018.06.093 Krisnandhika B, Faqih A, Pumamasari P D, Kusumoputro B. Emotion recognition system based on EEG signals using relative wavelet energy features and a modified radial basis function neural networks. In: Proceedings of the 2017 International Conference on Consumer Electronics and Devices (ICCED). London, UK: IEEE, 2017. 50−54 孙颖, 马江河, 张雪英. 结合非线性全局特征和谱特征的脑电情感识别. 计算机工程与应用, 2018, 54(17): 116-121 doi: 10.3778/j.issn.1002-8331.1803-0027

Sun Ying, Ma Jiang-He, Zhang Xue-Ying. EEG emotion recognition based on nonlinear global features and spectral feature. Computer Engineering and Applications, 2018, 54(17): 116-121 doi: 10.3778/j.issn.1002-8331.1803-0027 李幼军, 黄佳进, 王海渊, 钟宁. 基于SAE 和LSTM RNN的多模态生理信号融合和情感识别研究. 通信学报, 2017, 38(12): 109-120 doi: 10.11959/j.issn.1000-436x.2017294

Li You-Jun, Huang Jia-Jin, Wang Hai-Yuan, Zhong Ning. Study of emotion recognition based on fusion multi-modal bio-signal with SAE and LSTM recurrent neural network. Journal on Communications, 2017, 38(12): 109-120 doi: 10.11959/j.issn.1000-436x.2017294

地址：北京中关村东路95号 邮政编码：100190 E-mail： [email protected]

电话：010-82544677 (日常咨询和稿件处理)， 010-82544653(费用管理、寄刊)

北京仁和汇智信息技术有限公司 开发 技术支持： [email protected]