Classification of Mental Tasks from EEG Signals Using Spectral Analysis, PCA and SVM

摘要

Signals provided by the ElectroEncephaloGraphy (EEG) are widely usedin Brain-Computer Interface (BCI) applications. They can be further analyzed andused for thinking activity recognition. In this paper we proposed an algorithm thatis able to recognize five mental tasks using 6 channel EEG data. The main idea is toseparate the raw EEG signals into several frames and compute their spectrums.Next, a second-order derivative of Gaussian is applied to extract features and anoptimum Gaussian kernel parameters grid search is performed with the help ofcross-validation. The extracted features are further reduced by PrincipalComponent Analysis. The processed data is utilized to train SVM classifier which isused for mental tasks recognition afterwards. The performance of the algorithm isestimated on publically available dataset. In terms of 5 folds cross-validation weobtained an average of 82.7% recognition rate (accuracy). Additional experimentswere conducted using leave-one-out cross-validation where 67.2% correctclassification was reported. Comparison to several state-of-the art methods revealsthe advantages of the proposed algorithm.