针对无效采样的脑电信号插值研究
首发时间:2018-05-25
摘要:对信号采样进行插值是处理采样间断的常用方法,这种方法或许也能够消除脑电数据中无效采样的影响,使脑电识别顺利进行,本文就该问题进行了研究。本文以国际脑机接口竞赛中多个关于两类运动想象的脑电数据集为素材,利用小波变换、滑动平均等方法提取多种脑电信号特征,验证了3种基本的插值方式。研究得出,插值操作在一定程度上适用于解决由无效采样造成的脑电特征空缺问题,在应用插值操作时,除了插值方式的选择,还应注意无效采样占比、特征提取方式等因素的影响。插值操作可使样本完备,使脑电识别转为正常状态。
For information in English, please click here
Research on the Interpolation against Void Sampling in EEG
Abstract:Interpolation is a commonly used method in signal sampling for dealing with sampling discontinuities. This method may be able to eliminate the effect of void sampling in EEG data and make EEG identification proceed smoothly. In the paper, the two-type motor imagery EEG data sets of the international BCI competitions was used as EEG samples, the method of wavelet transform and moving average was used to extract various EEG features, and then 3 basic styles of interpolation were tested. The research indicated that the method of interpolation was suitable for solving the problem of EEG feature vacancy caused by void sampling to a certain extent. Besides the choice of interpolation styles, some factors such as the proportion of void sampling and the style of feature extraction etc should be noticed when applying interpolation. Interpolation can make sample complete and change EEG identification into normal state.
Keywords: BEI (Biotic Electric Interface) EEG (electroencephalogram) interpolation sampling CR (conformity rate)
基金:
引用
No.****
动态公开评议
共计0人参与
勘误表
针对无效采样的脑电信号插值研究
评论
全部评论0/1000