Study on mental fatigue detection of crane operators based on frequency and channel attention spiking neural network with multivariate variational mode decomposition in real construction environments.

To address the issues of strong noise interference and low detection accuracy of mental fatigue in tower crane operators under complex construction site environments, this study proposes a novel mental fatigue recognition method based on a frequency and channel attention spiking neural network with multivariate variational mode decomposition (MVMD-FCASNN). The method employs multivariate variational mode decomposition (MVMD) to jointly decompose multichannel electroencephalogram (EEG) signals and extract frequency-aligned intrinsic mode functions (IMFs), thereby enhancing feature separability and robustness against interference. An IMF attention mechanism is designed to adaptively evaluate the contribution of distinct frequency bands, while a channel attention module emphasizes critical brain regions, particularly the frontal and central areas, to strengthen spatial representation. The extracted features are further processed using a spiking neural network (SNN) to preserve the temporal dynamics of EEG signals and improve performance under noisy conditions. Experimental results demonstrate that the proposed MVMD-FCASNN model achieves an accuracy of 98.81% in mental fatigue classification tasks and maintains strong performance even under high noise conditions (-6 dB), significantly outperforming traditional methods. This approach offers a reliable and efficient solution for real-time mental fatigue monitoring in high-noise construction scenarios.