论文部分内容阅读
针对基于EEG的脑—机接口(BCI)实验数据分布不明朗的特点,双滤波模式(DFP)算法利用样本模式相似性来优化BCI的分类特征——运动相关电位(MRPs)特征的空间(即电极位置)和时间投影方向,使得映射后异类样本模式差异性与同类相似性的比值最大化。该算法考虑MRPs特征对时间、空间的敏感性,并以自适应的方式挖掘它们适合分类的信息;优化时不需要进行样本数据分布假设,符合BCI数据特点。最后,DFP算法对BCI competitionⅠ、Ⅱ两组数据进行实验,识别效果均高于相关比赛的最好成绩,这表明DFP算法能有效提取MRPs特征。
In view of the uncertain EEG-based BCI experimental data distribution, the DFP algorithm uses sample pattern similarity to optimize the classification features of BCI - the space of motion-dependent potential (MRPs) features Electrode position) and time projection direction, the ratio of the difference between the heterogeneous sample pattern and the similar similarity after mapping is maximized. The algorithm takes into account the temporal and spatial sensitivity of the MRPs features and mining their information suitable for classification in an adaptive manner. The optimization does not require the assumption of distribution of the sample data, which is consistent with the characteristics of the BCI data. Finally, DFP algorithm is used to test the data of BCI competition Ⅰ and Ⅱ, and the recognition effect is higher than the best result of the relevant competition, indicating that the DFP algorithm can effectively extract MRPs features.