Emotion Recognition from EEG Signals Using Multidimensional Information in EMD Domain.

Area of Science:

• Neuroscience
• Signal Processing
• Machine Learning

Background:

• Emotion recognition from electroencephalogram (EEG) signals is crucial for brain-computer interfaces.
• Traditional feature extraction methods may not fully capture the complex dynamics of EEG signals during emotional states.

Purpose of the Study:

• To introduce a novel feature extraction method for emotion recognition using Empirical Mode Decomposition (EMD).
• To evaluate the effectiveness of EMD-derived features in improving EEG-based emotion recognition accuracy.

Main Methods:

• EEG signals are decomposed into Intrinsic Mode Functions (IMFs) using EMD.
• Multidimensional features, including time series differences, phase differences, and normalized energy of IMFs, are extracted.
• Performance is validated on the publicly available DEAP dataset.

Main Results:

• The proposed EMD-based features demonstrate effectiveness in emotion recognition.
• High-frequency component IMF1 shows a significant impact on detecting different emotional states.
• The method achieves improved classification accuracy compared to Fractal Dimension (FD), Sample Entropy, Differential Entropy, and Discrete Wavelet Transform (DWT).

Conclusions:

• EMD is a powerful tool for extracting informative features from EEG signals for emotion recognition.
• The proposed method offers a promising advancement in the field of affective computing.
• Further analysis of informative electrodes using EMD strategy can refine emotion recognition systems.