|
|||||
|
|
| 基于运动想象的脑电信号特征提取与分类 | |
| 引用本文： | 李丽君,黄思娟,吴效明,熊冬生. 基于运动想象的脑电信号特征提取与分类[J]. 医疗卫生装备, 2011, 32(1): 16-17,25 |
| 作者姓名： | 李丽君 黄思娟 吴效明 熊冬生 |
| 作者单位： | 华南理工大学生物科学与工程学院,广州,515006 |
| 基金项目： | 广东省科技技术项目(2009B030801004); 广州市天河区科技计划项目(096G135) |
| 摘 要： | 目的：以在已知类别的2种运动想象任务下采集的EEG信号为训练样本,识别测试样本中的运动想象任务。方法：在频域范围内,采用AR模型功率谱估计法所得C3、C4通道的功率谱密度,确定ERD/ERS较明显的频率范围;在时域范围内,比较C3、C4通道信号的能量差异,确定ERD/ERS较明显的时间段。采用带通滤波和小波包分析的方法提取训练集想象运动过程中ERD/ERS生理现象较明显的节律信号,分别采用线性分类器、支持向量机（SVM）实现测试集运动想象脑电数据的分类。结果：分类最佳正确率为87.14%。结论：小波包分析法能够较准确地提取想象左、右手运动的脑电信号的本质特征,结合支持向量机实现较好的抗干扰能力和分类性能。 |
| 关 键 词： | 脑机接口 脑电信号 小波包分析 支持向量机 |
Feature Extraction and Classification of Imaginary Movements in EEG |
|
| LI Li-jun,HUANG Si-juan,WU Xiao-ming,XIONG Dong-sheng. Feature Extraction and Classification of Imaginary Movements in EEG[J]. Chinese Medical Equipment Journal, 2011, 32(1): 16-17,25 | |
| Authors: | LI Li-jun HUANG Si-juan WU Xiao-ming XIONG Dong-sheng |
| Affiliation: | LI Li-jun,HUANG Si-juan,WU Xiao-ming,XIONG Dong-sheng(School of Bioscience & Bioengineering,South China University of Technology,Guangzhou 510006,China) |
| Abstract: | Objective To recognize the imaginary movement profiles in the test samples using the EEG signals with two kinds of known imaginary movements as the training samples.Methods In the frequency range,AR model spectrum estimation method was used to obtain the power spectral density from C3,C4 channel to determine the observably ERD/ERS frequency range;in the time domain,the energy difference between C3 and C4 channel was found out to determine the observably ERD/ERS time section.Band-pass filter and wavelet pack... |
| Keywords: | Brain-computer interface EEG signal wavelet package analysis support vector machine |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ | |