Single-Trial Event-Related Potentials Classification via a Discriminative Dictionary Learning Scheme
Due to the better performances with large number of training samples, algorithms based on sparse representation have received more and more attentions in single-trial event-related potentials classification. Considered the burden from repeating psychological experiments, the classification with less training samples is still a challenge in both cognitive science and pattern recognition. In this paper, a discriminative dictionary learning based scheme is utilized to single-trial ERPs classification, in order to enhance the performance when the training sample size is small. After preprocessing, wavelet is employed to remove the strong background noise at first, and then a sparse representation recognition method based on discriminative dictionary learning, called D-KSVD, is applied to perform the classification on each testing trial. Experiments on ERPs epochs from risk decision test have demonstrated that proposed approach outperforms than existing sparse representation classifier when the training samples decrease dramatically.
Keywordssingle-trial ERPs classification sparse representation less training samples dictionary learning
Unable to display preview. Download preview PDF.
- 1.Blankertz, B., Curio, G., Muller, K.R.: Classifying Single Trial EEG: Towards Brain Computer Interfacing. NIPS 1, 157–164 (2002)Google Scholar
- 5.Tomioka, R., Aihara, K., Muller, K.R.: Logistic regression for single trial EEG classification. NIPS 19, 1377–1384 (2007)Google Scholar
- 9.Shin, Y., Lee, S., Lee, J., Lee, H.N.: Sparse representation-based classification scheme for motor imagery-based brain computer interface systems. Journal of Neural Engineering 9(5), 056002 (2012)Google Scholar
- 10.Zhang, Q., Li, B.: Discriminative k-svd for dictionary learning in face recognition. In: 2010 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 2691–2698. IEEE Press (2010)Google Scholar