神经网络在两类运动想象脑电识别中特征综合能力的检验
首发时间:2018-04-20
摘要:神经网络在脑电识别中对各种复合特征卓越的综合分析能力是该方法的优势,本文通过检验,直观地展示了该方法的这种性能特点,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,利用小波变换方法提取脑电信号的特征,以神经网络方法进行脑电识别,对识别结果加以分析比对,展示出神经网络对于复合脑电特征识别的普遍优势。本文直观且具体的分析也有助于对各种脑电特征的考察及选取,以期充分发挥神经网络在脑电识别中的优势。
For information in English, please click here
Synthesizing Ability Tests to Neural Network in EEG Identification about Two Types of Motor Imagery
Abstract:Excellent synthesizing ability of the neural network to various composite features in EEG identification is the advantage of this method. The paper shows such performance characteristic of this method intuitively through a series of tests. Based on the two-type motor imagery EEG data sets of the international BCI competitions, the method of wavelet transform was used to extract EEG features, and the neural network was used in EEG identification. Through comparing and analyzing to the identification results, general advantage of neural network to composite EEG features identification was exhibited. The intuitive and concrete analysis of the paper can also be helpful to inspecting and selecting various EEG features, in order to make this method give full play in EEG identification.
Keywords: BEI (Biotic Electric Interface) EEG (electroencephalogram) neural network wavelet transform conformity rate
基金:
引用
No.****
动态公开评议
共计0人参与
勘误表
神经网络在两类运动想象脑电识别中特征综合能力的检验
评论
全部评论0/1000