摘要
基于瞬态视觉诱发电位的研究是脑机接口研究中的一种方法,其核心在于瞬态视觉诱发电位的识别算法研究。采用累加平均和小波分解滤波从强噪声背景下提取微弱的视觉诱发电位,采用主成分分析提取诱发电位的特征,用K近邻算法对得到的特征信号进行模式识别。采集三名受试者的脑电数据作为处理对象,识别准确率可以达到95%。实验结果表明:该方法可以比较准确地识别瞬态视觉诱发电位。
The study based on transient visual evoked potentials is a kind of way to study brain-computer interface,and the key is the study on recognition algorithm of transient visual evoked potential.The accumulation averaging and wavelet filter method are used to extract VEP signal from the background full of noise.PCA are used to extract the feature of VEP signal.KNN is used to recognize the feature signal.The correct recognition rate reaches 95 % when EEG come from three people.The experimental results show that the method can recognize transient visual evoked potentials more correctly.
出处
《传感器与微系统》
CSCD
北大核心
2012年第4期47-49,共3页
Transducer and Microsystem Technologies
基金
重庆市科技攻关计划资助项目(CSTC
2009AC5023)
关键词
瞬态视觉诱发电位
小波变换
主成分分析
K近邻算法
transient visual evoked potential(TSVEP)
wavelet transform
principal component analysis(PCA)
K-nearest neighbor(KNN)algorithm