Improving underwater acoustic classification through learnable Gabor filter convolution and attention mechanisms

Remotely detecting and classifying underwater acoustic targets is critical for environmental monitoring and defence. However, the complexity of ship-radiated and environmental noise poses significant challenges for accurate signal processing. While recent advancements in machine learning have improved classification accuracy, limited dataset availability and a lack of standardised experimentation hinder reliable performance assessment and model development. This paper introduces GSE ResNeXt, a deep learning architecture integrating learnable Gabor convolutional layers with a ResNeXt backbone enhanced by squeeze-and-excitation attention. The Gabor filters serve as two-dimensional adaptive band-pass filters, extending the feature channel representation. Its combination with channel attention improves training stability and convergence while enhancing the model’s ability to extract discriminative features. The model is evaluated using three training-test split strategies that reflect increasingly complex classification tasks, demonstrating how evaluation design influences performance through factors such as data leakage, temporal separation, and taxonomy. Results show that GSE ResNeXt consistently outperforms baseline models in terms of classification performance. Regarding stability and convergence, adding Gabor convolutions to the initial layers of the model reduced training time by up to 41%. During the evaluation of training-testing splits, temporal separation between subsets significantly affected performance, proving more influential than training data volume. These findings suggest that frequency-selective feature extraction can improve optimisation behaviour and classification performance in underwater acoustic classification. Furthermore, the analysis highlights the importance of temporal separation and dataset composition when evaluating model performance. Future developments should focus on mitigating environmental effects on input signals.