EEG-signal-processing-脑电信号处理方法算法.pptx

EEG SIGNAL PROCESSING,第一页，共二十七页。,EEG signal modelling,1,Available features,2,Classification algorithms,3,Independent Component Analysis,4,Content,Sparse Representation,5,第二页，共二十七页。,1,EEG signal modelling,Bioelectricity,1,Signal generation system,2,第三页，共二十七页。,bioelectricity,Signal generation system,Excitation model,第四页，共二十七页。,signal generation system,bioelectricity,Linear Model,第五页，共二十七页。,signal generation system,bioelectricity,Nonlinear Model,第六页，共二十七页。,2,Available features,Basic features,1,Modern methods,2,第七页，共二十七页。,Temporal Analysis Signal Segmentation:label the EEG signals by segments of similar characteristics.,basic features,Modern methods,第八页，共二十七页。,Temporal Criteria,basic features,Modern methods,第九页，共二十七页。,Frequency AnalysisSuboptimal DFT,DCT,DWT;Optimal KLT(Karhunen-Love)Demerits:complete statistical information,no fast calculation.,basic features,Modern methods,第十页，共二十七页。,Signal Parameter Estimation AR model:Merits:Outperform DFT in frequency accuracy.Demerits:suffer from poor estimation of parameters.Improvements:accurate order&coefficients.,modern methods,Basic features,第十一页，共二十七页。,AR coefficients estimation methodsYule-Walker aryule(x,p)Merits:Toeplitz matrix Levinson-Durbin,fastest!Demerits:with window bad resolution of PSD,modern methods,Basic features,第十二页，共二十七页。,AR coefficients estimation methodsCovariance method arcov(x,p),armcov(x,p)Merits:without window good resolution of PSD Demerits:slowBurg arburg(x,p)Merits:accurate approximation of PSD Demerits:line skewing&splitting,modern methods,Basic features,第十三页，共二十七页。,modern methods,Basic features,Comparison,第十四页，共二十七页。,Principal Component AnalysisUse same concept as SVDDecompose data into uncorrelated orthogonal componentsAutocorrelation matrix is diagonalizedEach eigenvector represents a principal componentApplication decomposition,classification,filtering,denoising,whitening.,modern methods,Basic features,第十五页，共二十七页。,3,Sparse Representation,Sparse Approximation,1,Sparse Decomposition,2,第十六页，共二十七页。,Over-complete dictionary atomsHilbert space:Signal:Error:“Sparse:lN,satisfy limited error.,sparse approximation,Sparse decomposition,第十七页，共二十七页。,Major algorithms:Basic Pursuit,Matching Pursuits,OMPMatching Pursuits(MP):1st:kth:,sparse decomposition,Sparse approximation,与 正交,第十八页，共二十七页。,K-SVD:training dictionaryPotential applications for EEG:Coefficients featuresERP detectionAbnormal EEG detectionClassification of different status of EEG,sparse decomposition,Sparse approximation,第十九页，共二十七页。,4,Classification algorithms,Common methods,1,第二十页，共二十七页。,Nave BayesLDA:Linear Discriminant AnalysisHMM:Hidden Markov ModellingSVM:Support Vector MachineK-meansANNs:Artificial Neural NetworksFuzzy Logic,Common methods,第二十一页，共二十七页。,5,Independent Component Analysis,ICA approaches,1,Application,2,第二十二页，共二十七页。,Independent Component Analysis,ica approaches,applications,第二十三页，共二十七页。,ica approaches,applications,ICA approaches:Factorizing the joint PDF into its marginal PDFsDecorrelating signals through timeEliminating temporal cross-correlation function,第二十四页，共二十七页。,BSS:Blind Source SeparationNormal brain rhythms,event-related sources Artefacts eye movement&blinking,swallow,applications,Ica approaches,第二十五页，共二十七页。,THANKS!,第二十六页，共二十七页。,内容总结,EEG SIGNAL PROCESSING。Optimal KLT(Karhunen-Love)。Yule-Walker aryule(x,p)。Covariance method arcov(x,p),armcov(x,p)。Burg arburg(x,p)。Matching Pursuits(MP):。K-SVD:training dictionary。THANKS,第二十七页，共二十七页。,