Volume 32, Issue 1 (2014)                   jmciri 2014, 32(1): 27-40 | Back to browse issues page

XML Persian Abstract Print


Abstract:   (3523 Views)

Introduction: According to role of emotions in human’s life, if they can be recognized contemporary with facial expression by EEG
signals, it can discriminate between real abd artificial emotions. This ability can be applied in lie detector and be also helpful to patients that able to feel emotions but can’t show it with facial expression.

Methods: In this study, some emotion induction experiments were designed. But due to lack of emotion induction tools for doing such a research at Iranian culture, in addition to recognizing emotions, the second goal was to collect an emotion induction anthology or
collection based on Iranian culture. To achieve these goals, 24 subjects including 16 males and 8 famales were under study by International Affective Picture System (IAPS) and movies with Iranian culture.

Results: After reviewing various methods, fractal dimension and the AdaBoost classifier were selected as the best methods. 92% average accuracy was achieved from movies with Iranian culture and 79.6% average accracy was achieved for removing confounding effect of gender difference, emotion recognition was not performed separately for males and females and both sexes were included. 

Conclusion: Despite gender differences in participants; the results show that recognition methods are very strong. Also the results show that culturally related movies get higher accuracy in comparison with IAPS. So using EEG signals along and concomitant by facial
expressions can solve the problem of hiding emotions from facial
images.

Keywords: Emotion recognition, EEG signals, Facial expressions,
International Affective Picture System, Fractal dimension.

Full-Text [PDF 115 kb]   (6157 Downloads)    
Type of Study: Research | Subject: General

Rights and permissions
Creative Commons License This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.