基于多导联脑电时空信息的情感分类研究  被引量:2

Research for Emotion Classification Based on Temporal and Spatial Information of Multi-lead Electroencephalogram Signals

在线阅读下载全文

作  者:李铭 乔晓艳[1] LI Ming;QIAO Xiaoyan(College of Physics and Electronics Engineering,Shanxi University,Taiyuan 030006,China)

机构地区:[1]山西大学物理电子工程学院,山西太原030006

出  处:《山西大学学报(自然科学版)》2023年第1期167-175,共9页Journal of Shanxi University(Natural Science Edition)

基  金:山西省回国留学人员科研资助项目(2020009);太原市小店区产学研合作科技专项(201906)。

摘  要:为探究脑电空间关联信息与不同情感状态之间对应关系,基于SEED(SJTU Emotion EEG Dataset)情感脑电数据集,计算实验采集的不同导联脑电的皮尔逊相关系数,通过小波变换获取脑电导联之间的小波相干系数,利用Hilbert变换提取各个导联脑电的瞬时相位,计算脑电相位同步指数。然后分别将皮尔逊相关系数、小波相干系数、相位同步指数作为特征,采用支持向量机分类器实现正性、负性、中性三种情感状态的有效分类。仿真结果表明,脑电的空间关联特征用于情感识别是有效的,可以达到91.5%的情感识别精度;利用脑电微分熵的皮尔逊相关系数获得了93.7%的平均分类准确率;并且脑电γ节律相比α、β节律更有利于情感识别。该研究可以应用于情感脑机接口系统。To explore the correlation between the spatial association of Electroencephalogram(EEG) and the different emotional state, based on SEED dataset, Pearson correlation coefficients of the different leads are calculated using EEG signals collected in the experiment, and the wavelet coherence coefficients between the EEG lead signals are obtained by wavelet transform, then EEG phase synchronization index is extracted based on the Hilbert transform so as to calculate EEG instantaneous phase in each lead.These EEG features, including Pearson correlation coefficient, wavelet coherence coefficient, and phase synchronization index, are applied to achieve the classification of positive, negative, and neutral emotional states by using the Support Vector Machine. The simulation results show that the spatial correlation feature of EEG is effective for emotion recognition, and the accuracy of emotion recognition can reach to 91.5%;the Pearson correlation coefficient of EEG differential entropy has achieved an average classification accuracy of 93.7%;and EEG γ rhythm is more conducive to classifying emotion than α and β rhythms. The research can be applied to the emotional brain-computer interface system.

关 键 词:脑电情感识别 皮尔逊相关系数 小波相干系数 HILBERT变换 相位同步指数 

分 类 号:TP18[自动化与计算机技术—控制理论与控制工程]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象