静息脑电信号频域不对称指数特征识别算法

Frequency domain asymmetric exponential feature recognition algorithm for resting eeg signals

以实验的方式针对静息脑电信号的频域不对称指数(SASI)进行研究,通过图片诱发被实验者产生正性情绪、中性情绪和负性情绪,并将其脑电信号采集到一起,对采集到的脑电信号的后期正电位段进行处理分析,得到不同脑区在处于正性情绪、中性情绪和负性情绪时产生的不对称指数。实验研究结果表明,相对于中性情绪,实验者处于负性情绪时的静息脑电信号SASI值更大,实验者处于正性情绪时则更小,这一点在实验者的左颞区、右颞区和枕叶区有明显的表现。该研究对于提高识别情绪特征的准确率有重要意义。

The frequency domain asymmetry index(SASI)of resting EEG signals was studied experimentally.The positive,neutral and negative emotions of the subjects were induced by pictures,and the EEG signals were collected together.The positive potential segments of the collected EEG signals were processed and analyzed,and the result was No.Get the asymmetry index generated by different brain regions when they are in different emotions which are Positive,neutral or negative emotions.The experimental results show that the frequency domain asymmetry index of resting EEG signal in negative emotions is larger than that in neutral emotions,and the frequency domain asymmetry index of resting EEG signal in positive emotions is smaller,which is evident in left temporal area,right temporal area and occipital lobe of the experimenters.This study is of great significance to improve the accuracy of emotion recognition.