摘要
提出了基于信号频带能量和相同步作为脑电特征向量,实现对左右手运动意识任务的分类方法.用线性判别式算法对左右手运动想象脑电模式进行识别,识别正确率最高达到了86.43%,与只用特定频带能量作为脑电特征分类结果相比,效果更好.为大脑运动意识任务的分类提供了新思路.
A method for single trial on-line classification of imaginary hand movements, based on band power and phase synchronization derived from EEG, is proposed. The event-related EEG patterns during left and right hand motor imagery are identified by using the linear discriminant algorithm. The results show that the method is effective and the correct rate of classification is up to 86.43 %. It might provide a new way for the classification of mental tasks.
出处
《甘肃科学学报》
2008年第2期75-78,共4页
Journal of Gansu Sciences
基金
甘肃省高等学校研究生导师科研项目(0710-05)
兰州理工大学博士基金(SB03200702)
关键词
脑电信号
频带能量
相同步
特征提取
分类
EEG
band power
phase synchronization
feature extraction
classification