摘要：

在由容积传导采集的脑电数据中,可用于识别的信号非常模糊。为此,对三通道采集的运动想象脑电数据进行分析,融合多个识别结果以提高识别效果。预处理三通道采集的脑电数据,分别计算每个通道的功率谱,提取运动想象相关的功率谱值作为特征值,应用线性识别方法及Choquet 模糊积分对得到的多个结果进行融合。使用 2003 年国际BCI 竞赛数据和实验室测得的数据验证融合效果,结果显示融合后的识别准确率明显高于单一识别器。

关键词: 脑电数据, 信息融合, 线性识别分析, Choquet 模糊积分, 功率谱, 运动想象

Abstract:

Due to the volume conduction multi-channel Electroencephalogram(EEG)recordings give a rather blurred image of brain activity. Three-channel motor imagery EEG is analyzed. The result of EEG recognition is not satisfactory. It studies multiple classifier fusion to improve the motor imagery EEG classification accuracy,extracts feature from power spectrum calculated from the EEG, and designs classifiers based on the well-known Linear Discriminant Analysis(LDA)method. The fusion of the individual classifiers is realized by means of the Choquet fuzzy integral. BCI competition 2003 dataset III is used to validate fusion method. It demonstrates that the proposed method comes with better performance when compared with single techniques,and shows the effectiveness of the proposed method for dealing with EEG.

Key words: Electroencephalogram ( EEG) data, information fusion, Linear Discriminant Analysis ( LDA), Choquet fuzzy integral, power spectrum, motor imagery

中图分类号: 

• TP391