计算机科学与探索 ›› 2017, Vol. 11 ›› Issue (5): 794-801.DOI: 10.3778/j.issn.1673-9418.1603055

• 人工智能与模式识别 • 上一篇    下一篇

脑电信号与个人情绪状态关联性分析研究

陈  萌+,李幼军,刘  岩   

  1. 北京工业大学 电子信息与控制工程学院,北京 100124
  • 出版日期:2017-05-01 发布日期:2017-05-04

Analysis of Correlation Between EEG Signal and Personal Emotional State

CHEN Meng+, LI Youjun, LIU Yan   

  1. College of Electronic Information & Control Engineering, Beijing University of Technology, Beijing 100124, China
  • Online:2017-05-01 Published:2017-05-04

摘要: 人的情绪是人们对于客观事物是否满足自身需求而产生的一种综合状态,与生理信号有着密切的关联。对被试者心理状态剖面图(profile of moods states,POMS)的分量值和同时记录的个体静息态的脑电信号(electroencephalogram,EEG)特征值进行关联性分析研究。用小波变换对原始脑电信号进行预处理,脑电信号的特征值提取过程采用了经验模态分解(empirical mode decomposition,EMD)的方法,从预处理过的脑电信号中提取波动指数作为脑电特征值,随后将提取出的脑电特征值与POMS各分量值进行Pearson关联性分析。通过对8个被试者连续7天的POMS量表和脑电信号的记录与分析,得到脑电信号与情绪量表中的分量存在一定的正相关关联。

关键词: 心理状态剖面图(POMS)量表, 脑电信号(EEG), 经验模态分解(EMD), 皮尔森相关性分析

Abstract:  People's emotion is a kind of comprehensive state whether objective things meet people own needs is closely associated with the physiological signal. The subjects POMS (profile of moods states) scales and the EEG (electroencephalogram) signals of resting state at the same period are collected to study the correlation between them. Wavelet transform is used to preprocess the original EEG signal. EMD (empirical mode decomposition) method is adopted to decompose the denoised EEG into IMFs (intrinsic mode functions). Volatility index is extracted from the IMFs as a feature. The EEG features and POMS scales are analyzed by Pearson correlation analysis. The POMS scale and EEG signals are recorded and analyzed in 8 subjects for a period of 7 days. The experiment shows that there exists a strong correlation between the EEG signal and the related components of the personal emotion state.

Key words: profile of moods states (POMS) scale, electroencephalogram (EEG), empirical mode decomposition (EMD), Pearson correlation analysis