摘要
提出了一种基于Fisher判据的运动相关脑电特征优化选择的时频分析方法,为优化选择与运动意识最相关的有效脑电频率成分提供了理论依据.在此基础上,利用具有高时频分辨率的Morlet小波方法,提取与脑电特征优化过程最为匹配的特征信息,对4个受试者运动相关脑电模式进行分类,平均最大分类正确率达到87.95%,通过最大分类正确率、最大互信息两项评价指标比较,验证了基于匹配追踪和Fisher判据时频分析的特征优化方法对改善大脑运动意识任务分类性能的有效性.实验结果表明,该文所提出的方法可望应用于脑机接口系统的运动相关脑电特征优化和选择中.
A novel time frequency analysis method based on Fisher criterion for optimization of movement-related EEG power features is proposed to provide a theoretical guide for selecting the most relevant EEG frequency components. With Morlet wavelet filter to extract the optimized movement-related EEG features, two classes of EEG patterns for 4 subjects are discriminated and the average maximum classification accuracy reaches to 87.95%. By the two evaluation indexes, i.e. maximum classification accuracy and mutual information （MI）, the effectiveness of feature optimization based on time-frequency analysis of Fisher-ratio for improving classification performance is verified. The experimental results indicate the further applications of the propose method to the movement related EEG feature component selection and optimization.
出处
《西安交通大学学报》
EI
CAS
CSCD
北大核心
2008年第8期1026-1030,共5页
Journal of Xi'an Jiaotong University
基金
中国博士后科学基金资助项目(20070410380)
国家自然科学基金资助项目(30370395
30670534)
作者简介
裴晓梅（1970～），女．讲师．
