Emotion recognition from physiological signals using fusion of wavelet based features

Abstract : In this paper we propose a new system for human emotion recognition based on multi resolution analysis of physiological signals. In our study we have used four kinds of bio signals EMG, RESP, ECG and SC recorded at the University of Augsburg. Daubechies Symlet, Haar and Morlet wavelet transform were applied to analyze the non-stationary signals. Physiological features was extracted from the most relevant wavelet coefficients and the feature vectors obtained from each signal were combined using multimodal fusion technique to construct one feature vector for each emotion. A support vector machine (SVM) was adopted as a pattern classifier, an improved recognition accuracy of 95% was obtained and it clearly proves the performance of our new wavelet based approach in emotion recognition.
Document type :
Conference papers
Complete list of metadatas

https://hal.univ-lorraine.fr/hal-01785642
Contributor : Lcoms Ul <>
Submitted on : Friday, May 4, 2018 - 2:45:11 PM
Last modification on : Saturday, May 5, 2018 - 1:19:21 AM

Identifiers

Collections

Citation

Zied Guendil, Zied Lachiri, Choubeila Maaoui, Alain Pruski. Emotion recognition from physiological signals using fusion of wavelet based features. 2015 7th International Conference on Modelling, Identification and Control (ICMIC), Dec 2015, Sousse, Tunisia. ⟨10.1109/ICMIC.2015.7409485⟩. ⟨hal-01785642⟩

Share

Metrics

Record views

61