目录
摘要
基于瞬态P300成分在脑-机接口研究的中，通过P300成分信号的提取与识别产生脑-机接口控制信号。在强噪声环境下利用累加平均法和小波滤波提取微弱的P300成分。在小波变换域中计算特征向量，特征向量输入SVM分类识别，判断是否为P300成分。实验结果表明，小波分析法和累加平均法相结合可以有效的去除噪声得到P300成分信号；SVM分类器识别算法可以有效识别P300成分。
关键词：P300成分；脑-机接口；小波变换；SVM分类器
abstract
Based on the transient visual evoked potential in the study of brain computer interface, the control signal of brain computer interface was produced by the extraction and recognition of visual evoked potential signal. This experiment used averaging method and wavelet filter to extract weak visual evoked potential in the strong noise environment. Feature vectors are calculated in wavelet transform domain and they are input to SVM classification, to determine whether the signals are the visual evoked potential or not. The experimental results show that the combination of wavelet analysis and the cumulative average method can effectively remove the noise and get the visual evoked potential signal;.SVM classifier recognition algorithm can effectively identify the visual evoked potential.
Key words: visual evoked potential; brain computer interface; wavelet transform; SVM classifier
