Stabilizing underwater acoustic data generation with generative adversarial network

Abstract

Because of complex marine environments and scarce data, underwater acoustic target classification (UATC) is challenging. To improve model generalization ability, data augmentation methods, particularly data synthesis methods based on generative adversarial networks (GANs), are widely adopted. However, the training of GANs is usually slow and unstable. To address these issues, this paper proposes the adaptive stable deep convolutional GAN (AS-DCGAN). We introduce an adaptive controller that controls the learning progress based on the network training performance, thereby avoiding redundant training and accelerating the process. Additionally, we propose a progressive learning strategy that forces the network to gradually learn from low to high frequencies, stabilizing the training. We evaluated AS-DCGAN on two public datasets. The results show that our proposed method achieves state-of-the-art performance, with an accuracy of 81.14% on the DeepShip dataset and 86.11% on the ShipsEar dataset. Therefore, data augmentation and multimodel fusion methods can generate higher-quality data and effectively enhance the performance of UATC classification models.

Keywords

Generative adversarial networks / Deep learning / Data augmentation / Underwater target classification / Information and Computing Sciences / Artificial Intelligence and Image Processing

References

[1] AbdulZK, Al-TalabaniAK. Mel frequency cepstral coefficient and its applications: a review. IEEE Access, 2022, 10: 122136-122158.

[2] Aksüren IG, Hocaoǧlu AK (2022) Automatic target classification using underwater acoustic signals. In: 2022 30th Signal Processing and Communications Applications Conference (SIU). IEEE, pp 1–4

[3] Azimi-SadjadiMR, YaoD, HuangQ, DobeckGJ. Underwater target classification using wavelet packets and neural networks. IEEE Trans Neural Netw, 2000, 113784-794.

[4] Barratt S, Sharma R (2018) A note on the inception score. Preprint at arXiv:1801.01973

[5] BynagariNB. GANs trained by a two time-scale update rule converge to a local nash equilibrium. Asian J Appl Sci Eng, 2019, 8: 25-34.

[6] DubeySR, SinghSK, ChaudhuriBB. Activation functions in deep learning: a comprehensive survey and benchmark. Neurocomputing, 2022, 503: 92-108.

[7] IrfanM, ZhengJB, AliS, IqbalM, MasoodZ, HamidU. DeepShip: an underwater acoustic benchmark dataset and a separable convolution based autoencoder for classification. Expert Syst Appl, 2021, 183: 115270.

[8] Krichen M (2023) Generative adversarial networks. In: 2023 14th International Conference on Computing Communication and Networking Technologies (ICCCNT). IEEE, pp 1–7

[9] KrizhevskyA, SutskeverI, HintonGE. ImageNet classification with deep convolutional neural networks. Commun ACM, 2017, 60684-90.

[10] Ledig C, Theis L, Huszár F, Caballero J, Cunningham A, Acosta A et al (2017) Photo-realistic single image super-resolution using a generative adversarial network. In: 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). IEEE, pp 105–114

[11] LeiZF, LeiXJ, WangN, ZhangQY. Present status and challenges of underwater acoustic target recognition technology: a review. Front Phys, 2022, 10: 1044890.

[12] LiJ, SkinnerKA, EusticeRM, Johnson-RobersonM. WaterGAN: unsupervised generative network to enable real-time color correction of monocular underwater images. IEEE Robot Autom Lett, 2018, 31387-394

[13] LiL, LiXX, YinJW. Research on classification algorithm of ship radiated noise data based on generative adversarial network. J Electron Inf Technol, 2022, 4461974-1983

[14] Luo T, Ma Z, Xu ZQJ, Zhang YY (2019) Theory of the frequency principle for general deep neural networks. Preprint at arXiv:1906.09235

[15] Mehta S, Rastegari M (2021) MobileViT: light-weight, general-purpose, and mobile-friendly vision transformer. Preprint at arXiv:2110.02178

[16] Mirza M, Osindero S (2014) Conditional generative adversarial nets. Preprint at arXiv:1411.1784

[17] Radford A, Metz L, Chintala S (2015) Unsupervised representation learning with deep convolutional generative adversarial networks. Preprint at arXiv:1511.06434

[18] Santos-DomínguezD, Torres-GuijarroS, Cardenal-LópezA, Pena-GimenezA. ShipsEar: an underwater vessel noise database. Appl Acoust, 2016, 113: 64-69.

[19] TangN, WangXY, ZhouF, TangSY, LyuYH. Reparameterization causal convolutional network for automatic modulation classification. IEEE Trans Veh Technol, 2024, 7368576-8583.

[20] Wu H, Song QZ, Jin GH (2018) Deep learning based framework for underwater acoustic signal recognition and classification. In: Proceedings of the 2018 2nd International Conference on Computer Science and Artificial Intelligence. Association for Computing Machinery, pp 385–388

[21] XiongZY, ZhuXQ. Ship radiated-noise research based on the LOFAR spectrum and DEMON spectrum characteristics. J Ship Mech, 2007, 112300-306(in Chinese with English abstract)

[22] YinJW, ZhuGJ, HanX, GuoLX, LiL, GeW. Temporal correlation and message-passing-based sparse Bayesian learning channel estimation for underwater acoustic communications. IEEE J Ocean Eng, 2024, 492522-541.

[23] ZhouT, SiJK, WangLY, XuC, YuXY. Automatic detection of underwater small targets using forward-looking sonar images. IEEE Trans Geosci Remote Sens, 2022, 60: 1-12

RIGHTS & PERMISSIONS

The Author(s)