A Comparison of Data Augmentation Methods for Training Deep Neural Networks on Synthetic Aperture Sonar

Abstract:In this work we study Automatic Target Recognition (ATR) for Synthetic Aperture Sonar (SAS) data with a focus on deep neural networks (DNNs). The main challenge in training DNNs for SAS ATR arises from the limited quantity of labeled target examples, which arises due to the significant costs and time required to collect real-world SAS data. One successful general strategy for mitigating the problem of limited training data is augmentation, which generates additional synthetic training data by introducing realistic variations to available real-world data. Prior research has investigated a variety of augmentation strategies for SAS ATR, including conventional image augmentations (e.g., contrast changes, cropping) as well as physics-based augmentations, which are motivated the specific physics of SAS data. Building on prior work, we systematically compare many of these existing augmentation strategies for training DNNs for SAS ATR. We also investigate the impact of augmentation when combined with modern DNN architectures such as transformers. The results indicate that augmentation can improve target recognition accuracy, although benefits vary, and not all augmentations are beneficial.

Machine Learning (cs.LG)

arXiv:2607.23770 [cs.LG]

(or arXiv:2607.23770v1 [cs.LG] for this version)

https://doi.org/10.48550/arXiv.2607.23770 Focus to learn more arXiv-issued DOI via DataCite (pending registration)

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.