Related Experiment Videos
Eeg emotion recognition based on channel spatio-temporal multi-dimensional feature extraction network
Jingjie Yan1, Siya Zhao1, Jing Li1
1Jiangsu Key Laboratory of Intelligent Information Processing and Communication Technology, College of Telecommunications and Information Engineering, Nanjing University of Posts and Telecommunications, Nanjing, 210003 China.
Cognitive Neurodynamics
|July 28, 2026
Summary
This study introduces a novel network for recognizing emotions from electroencephalogram (EEG) signals. The Channel Spatio-temporal Multi-dimensional Feature Extraction Network (CSMFEN) effectively extracts complex emotional features for improved EEG emotion recognition.
Area of Science:
- Neuroscience
- Computer Science
- Artificial Intelligence
Background:
- Electroencephalogram (EEG) signals possess high data complexity, capturing temporal and spatial neuronal activity.
- Extracting deep emotional features from EEG signals is challenging due to this complexity.
Purpose of the Study:
- To propose a novel network, the Channel Spatio-temporal Multi-dimensional Feature Extraction Network (CSMFEN), for enhanced EEG emotion recognition.
- To effectively extract deeper emotional features from complex EEG signals.
Main Methods:
- CSMFEN learns features across channel, spatial, and temporal domains.
- Adaptive channel weighting enhances discriminative representations.
- Enhanced dynamic convolution captures inter-channel spatial relationships using attention mechanisms.
- Long Short-Term Memory (LSTM) networks model temporal dependencies.
Main Results:
- CSMFEN effectively learns discriminative representations from EEG signals.
- The proposed network achieves competitive emotion recognition performance on the SEED and DEAP datasets.
Conclusions:
- CSMFEN demonstrates significant potential for accurate EEG-based emotion recognition.
- The multi-dimensional feature extraction approach is effective for complex EEG data analysis.