Robust Identity Recognition Using Cognitive Task EEG Signals: An ICRM-ResNet Approach

In recent years, the application of deep learning techniques in electroencephalography (EEG) signal-based identity recognition has garnered widespread attention. However, noise interference and the mixing of source signals in EEG data remain prominent challenges. To address these issues, this paper proposes a robust identity recognition model, ICRM-ResNet, which combines Independent Component Analysis (ICA), Riemannian Manifold (RM) methods, and Residual Networks (ResNet) to enhance identity classification performance using EEG signals from cognitive tasks. By incorporating Riemannian manifold methods, the model strengthens feature extraction and overcomes the uncertainty in signal component ordering inherent in ICA, ensuring more stable signal representations. The model uses ResNet for deep learning to effectively classify these features. Experimental results on both a self-built dataset and a publicly available dataset validate the effectiveness and robustness of the proposed model in identity recognition tasks.