EEG-Based Driver Drowsiness Estimation Using Convolutional Neural Networks
This work addresses driver drowsiness detection, a safety-critical domain-specific problem, with an incremental improvement over existing methods.
The paper tackled driver drowsiness estimation from EEG signals by extending EEGNet to regression and using a spectral meta-learner, achieving improved performance with power spectral density inputs that reduce computational cost.
Deep learning, including convolutional neural networks (CNNs), has started finding applications in brain-computer interfaces (BCIs). However, so far most such approaches focused on BCI classification problems. This paper extends EEGNet, a 3-layer CNN model for BCI classification, to BCI regression, and also utilizes a novel spectral meta-learner for regression (SMLR) approach to aggregate multiple EEGNets for improved performance. Our model uses the power spectral density (PSD) of EEG signals as the input. Compared with raw EEG inputs, the PSD inputs can reduce the computational cost significantly, yet achieve much better regression performance. Experiments on driver drowsiness estimation from EEG signals demonstrate the outstanding performance of our approach.