Emotion recognition based on electroencephalogram signal analysis
Subject Areas :fatemeh adli 1 , hamid reza hosseinzadeh 2 * , Mohammad Mahdi Moradi 3
1 - Department of Electrical and Computer Engineering, North Tehran Branch, Islamic Azad University, Tehran
2 - Department of Electrical Engineering, North Tehran Branch, Islamic Azad University, Tehran, Iran.
3 - Faculty of Engineering, Shahid Chamran University, Kerman, Kerman, Iran
Keywords: Emotions, brain signal (EEG), feature extraction, classification,
Abstract :
In this research, a comprehensive system is introduced for emotion detection and classification based on nonlinear analysis of electroencephalogram (EEG) signals. For this purpose, data from the DEAP database was used, which includes signals recorded from 32 participants while watching music videos. The duration of each video was 120 minutes, and the individual's emotions were labeled separately for each minute. This labeling was performed by a psychologist through video analysis of the participants' facial expressions. In the next step, the EEG signals were subjected to noise removal and filtration processes. This process involved two stages of filtering: one in the frequency domain and the other in the time-frequency domain. A time-frequency domain filter with default coefficients suitable for removing noise from unknown sources was employed. After signal preparation, temporal, frequency, and trend-free nonlinear features were extracted. Finally, emotion classification was performed using a hybrid “Adaboost” classifier. The results showed that this method has a high capability in differentiating various emotions, achieving an accuracy of 97%, a sensitivity of 90%, and a specificity of 99%. This high performance confirms the superiority of the proposed method compared to similar articles. Overall, the use of hybrid classifiers significantly increases classification accuracy due to their ability to combine different features
[1] N. Ahmadzadeh Nobari Azar, N. Cavus, P. Esmaili, and others, “Detecting emotions through EEG signals based on modified convolutional fuzzy neural network,” Sci. Rep., vol. 14, no. 1, pp. 10371, May 2024.
[2] R. Sharma and H. K. Meena, “EmHM: A novel hybrid model for the emotion recognition based on EEG signals,” in 2023 19th IEEE International Colloquium on Signal Processing and Its Applications (CSPA), Malaysia, 2023, pp. 75-80.
[3] K. Singh, M. K. Ahirwal, and M. Pandey, “Quaternary classification of emotions based on electroencephalogram signals using hybrid deep learning model,” J. Ambient Intell. Humaniz. Comput., vol. 14, no. 1, pp. 2429–2441, 2023.
[4] M. Moradi, M. Fatehi, H. Masoumi, and M. Taghizadeh, “Transfer learning method for Sleep stages classification using different domain,” Asian J. Adv. Med. Sci., vol. 2, no. 3, pp. 21-25, 2020.
[5] L. Kanchi, C. Narra, G. Mantri, M. Palepogu, and C. Mettu, “Emotion Recognition from Brain EEG Signals,” Int. J. Res. Appl. Sci. Eng. Technol., vol. 12, pp. 2036-2043, 2024.
[6] M. Moradi, M. Fatehi, H. Masoumi, and M. Taghizadeh, “Deep neural network method for classification of sleep stages using spectrogram of signal based on transfer learning with different domain data,” Scientia Iranica, vol. 29, no. 4, pp. 1898-1903, 2022.
[7] J. Zhang, Y. Hao, X. Wen, C. Zhang, H. Deng, J. Zhao, and R. Cao, “Subject-Independent Emotion Recognition Based on EEG Frequency Band Features and Self-Adaptive Graph Construction,” Brain Sci., vol. 14, no. 3, pp. 271, Mar. 2024.
[8] Q. Li, Y. Liu, F. Yan, Q. Zhang, and C. Liu, “Emotion recognition based on multiple physiological signals,” Biomed. Signal Process. Control, vol. 85, 2023.
[9] M. Moradi, M. Fatehi, H. Masoumi, and M. Taghizadeh, “Sleep stages classification based on deep transfer learning method using PPG signal,” Signal Process. Renewable Energy, vol. 5, no. 2, pp. 53-60, 2020.
[10] H. Shariatmadar and S. Golnargesi, “Structural Control with Active Tuned Mass Damper Using Type 2 Interval Fuzzy Logic Controller for Seismic Excitation,” Sharif Civil Eng., vol. 21, pp. 21-32, 2015.
[11] M. Moradi, M. Fatehi, H. Masoumi, and M. Taghizadeh, “Adaptive neuro-fuzzy method for sleep stages detection by PPG signal,” J. Adv. Pharm. Educ. Res., vol. 10, no. S1, 2020.
[12] S. Bagherzadeh, A. Shalbaf, A. Shoeibi, M. Jafari, R.-S. Tan, and U. R. Acharya, “Developing an EEG-Based Emotion Recognition Using Ensemble Deep Learning Methods and Fusion of Brain Effective Connectivity Maps,” IEEE Access, vol. 12, pp. 50949-50965, 2024.
[13] M. H. Fatehi, M. Taghizadeh, M. M. Moradi, and P. Ravanbakhsh, “Diagnosis of Covid-19 using optimized convolutional neural network,” J. Artif. Intell. Electr. Eng., vol. 11, no. 42, pp. 48-54, 2021.
[14] F. Amini, A. Mohajeri, and M. Javanbakht, “Semi-active control of isolated and damaged structures using online damage detection,” Smart Mater. Struct., 2015.
[15] M. H. Fatehi, M. Khajooee, N. Adlband, and M. M. Moradi, “Detection of healthy and unhealthy ECG signal using optimized convolutional neural network,” J. Artif. Intell. Electr. Eng., vol. 11, no. 43, pp. 61-69, 2023.
[16] M. Mohebbi, H. Rasooli Dabbagh, and S. Moradpoor, “Seismic Vibration Control of Nonlinear Structures Using Semi-Active Synchronized Mass Damper Mechanism,” Amirkabir J. Civil Eng., vol. 46, no. 2, pp. 117-131, 2015.
[17] A. Preumont, M. Voltan, A. Sangiovanni, B. Mokrani, and D. Alaluf, “Active Tendon Control of Suspension Bridges,” Smart Struct. Syst., 2015.
[18] A. Saeidi and R. Vahdani, “Active vibration control of smart building frames by active tendon,” in 10th International Congress of Civil Engineering, 2015.
[19] W. Wei, “Targeted generative adversarial network (TWGAN-GP)-based emotion recognition of ECG signals,” E3S Web Conf., vol. 522, pp. 01042, 2024.
[20] F. Y. Cheng, H. Jiang, and K. Lou, Smart Structures: Innovative Systems for Seismic Response Control. Boca Raton, FL, USA: CRC Press, 2008.
[21] Z. Sun and S. S. Ge, Switched Linear Systems: Control and Design. London, UK: Springer, 2005