Abstract-Common spatial pattern （CSP） algorithm is a successful tool in feature estimate of brain-computer interface （BCI）. However, CSP is sensitive to outlier and may result in poor outcomes since it is based on...Abstract-Common spatial pattern （CSP） algorithm is a successful tool in feature estimate of brain-computer interface （BCI）. However, CSP is sensitive to outlier and may result in poor outcomes since it is based on pooling the covariance matrices of trials. In this paper, we propose a simple yet effective approach, named common spatial pattern ensemble （CSPE） classifier, to improve CSP performance. Through division of recording channels, multiple CSP filters are constructed. By projection, log-operation, and subtraction on the original signal, an ensemble classifier, majority voting, is achieved and outlier contaminations are alleviated. Experiment results demonstrate that the proposed CSPE classifier is robust to various artifacts and can achieve an average accuracy of 83.02%.展开更多
为了提高运动想象脑机接口系统的性能,基于脑-机接口竞赛数据比较了不同空间滤波器下获得的CSP特征,在支持向量(线性核和高斯核)(linear kernel support vector machine,LSVM and(gaussian kernel support vector machine,GSVM)),线性...为了提高运动想象脑机接口系统的性能,基于脑-机接口竞赛数据比较了不同空间滤波器下获得的CSP特征,在支持向量(线性核和高斯核)(linear kernel support vector machine,LSVM and(gaussian kernel support vector machine,GSVM)),线性判别分析(linear discrimination analysis,LDA),梯度提升决策树(gradient boosting descrision tree,GBDT)下的分类效果.比较结果表明,GBDT获得了比其它分类器更优的分类效果.进一步把最小绝对收缩和选择算法(the Least Absolute Shrinkage and Selectionator operator,LASSO)与以上四种分类器进行结合使用,发现其与GBDT结合使用后得到的平均分类准确率最高,比结合LSVM,GSVM和LDA分别提高了5.57%,4.57%,3.16%.展开更多
针对脑电信号具有非平稳性、非线性以及个体差异较大等特点而导致特征提取困难、分类准确率低的问题,提出一种基于LMD-CSP和随机森林(Random Forest,RF)的脑电信号分类方法。首先对脑电信号进行预处理,然后利用局部均值分解(Local Mean ...针对脑电信号具有非平稳性、非线性以及个体差异较大等特点而导致特征提取困难、分类准确率低的问题,提出一种基于LMD-CSP和随机森林(Random Forest,RF)的脑电信号分类方法。首先对脑电信号进行预处理,然后利用局部均值分解(Local Mean Decomposition,LMD)将预处理后的脑电信号分解为多个乘积函数(Product Function,PF)分量,并选出最具判别性的PF分量,再利用共空间模式(Common Spatial Pattern,CSP)分别对选出的PF分量进行特征提取,最后将得到的CSP特征输入随机森林分类器中进行分类识别。实验结果表明,该方法的平均分类准确率高达92.18%,远高于其他方法,证明了该方法的有效性。展开更多
为解决运动想象脑电(electroencephalogram, EEG)信号多分类传输速率慢、准确率低的问题,本研究利用“一对多”滤波组共空间模式(one vs rest filter bank common spatial pattern, OVR-FBCSP)和稀疏嵌入(sparse embeddings, SE)提出了...为解决运动想象脑电(electroencephalogram, EEG)信号多分类传输速率慢、准确率低的问题,本研究利用“一对多”滤波组共空间模式(one vs rest filter bank common spatial pattern, OVR-FBCSP)和稀疏嵌入(sparse embeddings, SE)提出了一种基于SE的多分类EEG信号分类方法。为降低多类任务特征提取的复杂度,提高分类效率,本方法首先采用OVR-FBCSP进行EEG信号特征提取;然后对其相应的标签矩阵进行低维嵌入,构建稀疏嵌入模型,分别计算训练和测试数据的嵌入矩阵;最后在嵌入空间中对训练和测试数据执行k最近邻(k-nearest neighbor, kNN)分类。本研究在BCI Competition IV-2a公开数据集进行了实验测试,并与其他分类方法进行了对比。实验结果表明,本研究方法拥有较高的分类准确率和较短的分析时间。展开更多
基金supported by the National Natural Science Foundation of China under Grant No. 30525030, 60701015, and 60736029.
文摘Abstract-Common spatial pattern （CSP） algorithm is a successful tool in feature estimate of brain-computer interface （BCI）. However, CSP is sensitive to outlier and may result in poor outcomes since it is based on pooling the covariance matrices of trials. In this paper, we propose a simple yet effective approach, named common spatial pattern ensemble （CSPE） classifier, to improve CSP performance. Through division of recording channels, multiple CSP filters are constructed. By projection, log-operation, and subtraction on the original signal, an ensemble classifier, majority voting, is achieved and outlier contaminations are alleviated. Experiment results demonstrate that the proposed CSPE classifier is robust to various artifacts and can achieve an average accuracy of 83.02%.
