摘要
脑电信号包含着大脑皮层活动的丰富信息,但同时也包含了大量的噪声,如何有效地从这些丰富的信息中提取有用特征,一直是该研究领域的热点问题。文中提出利用灰建模的方法进行脑电特征提取,具有一定的创新性。介绍了灰色建模机理及其在脑电特征提取中的应用,利用实测脑电信号建立了脑电GM(1,1)模型,并进行了模型参数估计和特征提取,用K近邻算法对所提取的特征参数进行了分类。分类结果表明,利用灰建模的方法进行脑电特征提取和分类的方法是可行、有效的,为脑电信号的特征提取提供了一种新的思路和方法。
Electroencephalogram （EEG） signals include plenfful information of the activities in the cortex of scalp caused by the physiological activities of the brain. But substantive noises are also interwoven. How to extract the brain feature efficiently is always a difficult problem in the research field. The principle and application of GM（1,1） in feature extraction of the EEG signal is described, the GM（1,1） of EEG signal is constructed using the actual data in this paper. The parameters of the grey model are estimated and the feature is extracted, then the feature parameters are classified with the KNN classifier. The result of classification indicates that it is feasible and available for EEG feature extraction with the grey modeling method and this method provides a new idea to the EEG signal processing and extraction.
出处
《计算机仿真》
CSCD
2006年第12期81-85,共5页
Computer Simulation
基金
国家自然基金(30470459)
西工大科技创新基金(M450212)
关键词
脑电
阿尔发波
参数
特征提取
EEG
Alpha rhythm
Parameter
Feature extraction
