文件名称:EEG-based-identification-method
介绍说明--下载内容均来自于网络,请自行研究使用
:基于脑电信号的身份识别是通过采集试验者的脑部信号来进行身份认证。对于同一个外部刺激或者主体在思考同一个
事件的时候,不同人的大脑所产生的认知脑电信号不同。选取与运动意识想象有关的电极后,分析不同个体在特定状况下脑
电的个体差异,采用以回归系数、能量谱密度、相同步、线性复杂度多种信号处理结合方法对运动想象脑电信号进行处理来
进行特征提取。组合多元特征向量并运用多层BP 神经网络对不同个体的脑电信号进行分类,并在不同的意识想象及不同数
据长度、不同的波段对试验者进行识别率验证分析。结果表明,不同运动想象的平均识别率均在80 以上,其中以想象舌头
运动的识别率较高,达到90.6 ,不同波段的识别率也表明意识想象的模式及相应波段对身份认别有较大的影响。-EEG-based identification to authenticate through the acquisition of experimental brain signals. For the same external stimuli, or the main thinking of the same
Event, different people s brains produced by cognitive EEG. Select imagine the electrodes and movement awareness, analysis of different individuals in a particular situation brain
Individual differences in electricity, the use of regression coefficients, the energy spectral density, phase synchronization, the linear complexity of a variety of signal processing combined with motor imagery EEG
For feature extraction. The combination of multiple feature vectors and the use of multi-layer BP neural network to classify the EEG signals of different individuals, and in a different sense of imagination and a different number of
Length, the band on the test to verify the analysis of the recognition rate. The results show that the average recognition rates of different motor imagery in more than 80 , which to imagine the tongue
The m
事件的时候,不同人的大脑所产生的认知脑电信号不同。选取与运动意识想象有关的电极后,分析不同个体在特定状况下脑
电的个体差异,采用以回归系数、能量谱密度、相同步、线性复杂度多种信号处理结合方法对运动想象脑电信号进行处理来
进行特征提取。组合多元特征向量并运用多层BP 神经网络对不同个体的脑电信号进行分类,并在不同的意识想象及不同数
据长度、不同的波段对试验者进行识别率验证分析。结果表明,不同运动想象的平均识别率均在80 以上,其中以想象舌头
运动的识别率较高,达到90.6 ,不同波段的识别率也表明意识想象的模式及相应波段对身份认别有较大的影响。-EEG-based identification to authenticate through the acquisition of experimental brain signals. For the same external stimuli, or the main thinking of the same
Event, different people s brains produced by cognitive EEG. Select imagine the electrodes and movement awareness, analysis of different individuals in a particular situation brain
Individual differences in electricity, the use of regression coefficients, the energy spectral density, phase synchronization, the linear complexity of a variety of signal processing combined with motor imagery EEG
For feature extraction. The combination of multiple feature vectors and the use of multi-layer BP neural network to classify the EEG signals of different individuals, and in a different sense of imagination and a different number of
Length, the band on the test to verify the analysis of the recognition rate. The results show that the average recognition rates of different motor imagery in more than 80 , which to imagine the tongue
The m
(系统自动生成,下载前可以参看下载内容)
下载文件列表
EEG-based identification method.pdf