2008 | OriginalPaper | Buchkapitel
The Extreme Energy Ratio Criterion for EEG Feature Extraction
verfasst von : Shiliang Sun
Erschienen in: Artificial Neural Networks - ICANN 2008
Verlag: Springer Berlin Heidelberg
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
Energy is an important feature for electroencephalogram (EEG) signal classification in brain computer interfaces (BCIs). It is not only physiologically rational but also empirically effective. This paper proposes extreme energy ratio (EER), a discriminative objective function to guide the process of spatially filtering EEG signals. The energy of the filtered EEG signals has the optimal discriminative capability under the EER criterion, and hence EER can as well be regarded as a feature extractor for distilling energy. The paper derives the solutions which optimize the EER criterion, shows the theoretical equivalence of EER to the existing method of common spatial patterns (CSP), and gives the computational savings EER makes in comparison with CSP. Two paradigms extending EER from binary classification to multi-class classification are also provided.