Multivariate Pseudo Wigner Ville Distribution based Emotion Detection from Electrical Activity of Brain


ÖZEL P., AKAN A., YILMAZ B.

10th International Conference on Electrical and Electronics Engineering (ELECO), Bursa, Türkiye, 30 Kasım - 02 Aralık 2017, ss.516-519 identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Basıldığı Şehir: Bursa
  • Basıldığı Ülke: Türkiye
  • Sayfa Sayıları: ss.516-519
  • Abdullah Gül Üniversitesi Adresli: Evet

Özet

Recently, there has been a rapid development in multivariate signal analysis to determine joint oscillations for multiple data channels. The emotion elicitation in an electroencephalogram (EEG) is a novel area to evaluate methods for emotional differences from brain signals. In this paper, utilizing the idea of joint instantaneous frequency of multivariate data, a multivariate extension of pseudo Wigner distribution is used for emotion recognition from EEG signals, in which different window sizes are employed to interpret the results. As a preliminary study, the best results are obtained as 90%, 75%, 65% in terms of valence, arousal and dominance scale respectively for larger window size.