基于约束独立分量分析的脑电特征提取

发布时间：2018-10-14 12:30

【摘要】：针对脑机接口(brain-computer interface,BCI)系统特征提取较慢的现状,提出基于约束独立分量分析(constrained independent component analysis,cICA)的P300特征提取方法.首先,针对各位P300实验被试,通过EEG图像研究其特有P300时域特性;然后,根据P300特性构建参考信号,并将参考信号与独立分量分析(independent component analysis,ICA)方法结合,基于64导联EEG,提取出与P300相关度最大的独立分量;最后,依据提取出的独立分量构造3维特征向量进行分类.实验采用线性分类器,针对BCI Competition II dataset IIb和BCI Competition III dataset II两组公共数据集进行了验证.结果表明,提出方法在3次叠加平均下识别正确率达67.1%,15次达95.2%,在相同实验条件下,分类时间也较其他方法缩短.
[Abstract]:A P300 feature extraction method based on constrained Independent component Analysis (constrained independent component analysis,cICA) is proposed to solve the problem of slow feature extraction in brain-computer interface (brain-computer interface,BCI) systems. First of all, the special P300 time-domain characteristics of P300 are studied by EEG images, and then the reference signal is constructed according to the P300 characteristics, and the reference signal is combined with the independent component analysis (independent component analysis,ICA) method. Based on 64-lead EEG, the independent components with the greatest correlation with P300 are extracted. Finally, 3D feature vectors are constructed according to the extracted independent components. The experiment uses linear classifier to verify two groups of common data sets: BCI Competition II dataset IIb and BCI Competition III dataset II. The results show that the accuracy rate of the proposed method is 67.1 times and 95.2 times on the average of three superpositions, and the classification time is shorter than that of other methods under the same experimental conditions.
【作者单位】：东北大学中荷生物医学与信息工程学院;东北大学机械工程与自动化学院;
【基金】：国家自然科学基金资助项目(61071057)
【分类号】：R318

【参考文献】