SVM-FastICA Based Detection Ensemble System of EEG

An EEG signal detection ensemble system to solve the low rate of vision detection is developed when analysis so many EEG signals. A novel FastICA method is presented, in which the independent component analysis approach is used to acquire the high order statistic information of EEG intrusion action mode and mapped the input mode space into the corresponding independent component space. Then the generalized maximal margin hyperplane is constructed in the independent component space using the support vector machine. Testing results show that the system integrates the features of FastICA and SVM to response real-time and lower the rate of false negative.