基于Hilbert-Huang变换的思维脑电分类技术研究

Research of Mental EEG Classification Based on Hilbert-Huang Transform

研究基于Hilbert-Huang变换的思维脑电分类方法。对思维脑电信号进行Hilbert-Huang时频预处理,经经验模式分解后,得到多阶固有模态分量,然后将经HHT变换后的时频窗口内的振幅标准差作为不同心理作业信号特征,再应用K-近邻对思维脑电信号进行分类决策。通过对Colorado州立大学EEG研究中心的三类思维脑电心理作业样本进行分类,平均正确率达到82.54%。经Hilbert-Huang变换得到的脑电信号特征,可以作为思维脑电分类的有效依据。

This paper studies the classification of EEG signals during mental tasks based on Hilbert-Huang transform method. The mental EEG signals was preprocessed by Hilbert-Huang transform in the time-frequency field. With empirical mode decomposition （EMD）, the data set was decomposed into a finite and often small number of intrinsic mode functions （IMF）. Using Hilbert transform to those IMF components yielded instantaneous amplitude and frequency. After obtaining features parameters of different mental tasks by using the amplitude standard deviation of time-frequency window, the pattern recognition method of K- neighbors was applied to optimal classification. The experiment picked three classes of mental EEG signals from the database of Colorado State University EEG research center. The mean classification accuracy rate was up to 82. 54%. The features of mental EEG got by Hilbert-Huang transform can effectively do automatic mental tasks classification.