EUVIMEDEUVIMED entdecken
Crossref

Self-supervised learning–based underwater acoustical signal classification via mask modeling

Kele Xu iD, Qisheng Xu, Kang You, Boqing Zhu iD, Ming Feng, Dawei Feng, Bo Liu

The Journal of the Acoustical Society of America · 01.07.2023 · 154 (1) · 5-15

DOI: 10.1121/10.0019937 ↗

Kurz gefasst

Fachartikel in der Zeitschrift The Journal of the Acoustical Society of America, online erschienen am 05.07.2023. Verfasst von Kele Xu und 6 weiteren Autor:innen. Beteiligte Einrichtung: National Key Laboratory of Parallel and Distributed Processing, Changsha, 410073, China. Die Arbeit wurde bisher 57-mal zitiert (Crossref) und nennt 54 Literaturangaben. Gefördert von: National Major Science and Technology Projects of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The classification of underwater acoustic signals has garnered a great deal of attention in recent years due to its potential applications in military and civilian contexts. While deep neural networks have emerged as the preferred method for this task, the representation of the signals plays a crucial role in determining the performance of the classification. However, the representation of underwater acoustic signals remains an under-explored area. In addition, the annotation of large-scale datasets for the training of deep networks is a challenging and expensive task. To tackle these challenges, we propose a novel self-supervised representation learning method for underwater acoustic signal classification. Our approach consists of two stages: a pretext learning stage using unlabeled data and a downstream fine-tuning stage using a small amount of labeled data. The pretext learning stage involves randomly masking the log Mel spectrogram and reconstructing the masked part using the Swin Transformer architecture. This allows us to learn a general representation of the acoustic signal. Our method achieves a classification accuracy of 80.22% on the DeepShip dataset, outperforming or matching previous competitive methods. Furthermore, our classification method demonstrates good performance in low signal-to-noise ratio or few-shot settings.

Autor:innen und Institutionen

  • National Key Laboratory of Parallel and Distributed Processing, Changsha, 410073, China.
  • Tongji University, Shanghai, China.

ORCID-Profile: Kele Xu, Boqing Zhu

Förderung

  • National Major Science and Technology Projects of China

Zitieren

APA

Xu, K., Xu, Q., You, K., Zhu, B., Feng, M., Feng, D., & Liu, B. (2023). Self-supervised learning–based underwater acoustical signal classification via mask modeling. The Journal of the Acoustical Society of America, 154(1), 5-15. https://doi.org/10.1121/10.0019937

Vancouver

Xu K, Xu Q, You K, Zhu B, Feng M, Feng D, et al. Self-supervised learning–based underwater acoustical signal classification via mask modeling. The Journal of the Acoustical Society of America. 2023;154(1):5-15. doi:10.1121/10.0019937

BibTeX
@article{Xu2023,
  author = {Xu, Kele and Xu, Qisheng and You, Kang and Zhu, Boqing and Feng, Ming and Feng, Dawei and Liu, Bo},
  title = {Self-supervised learning–based underwater acoustical signal classification via mask modeling},
  journal = {The Journal of the Acoustical Society of America},
  year = {2023},
  volume = {154},
  number = {1},
  pages = {5--15},
  doi = {10.1121/10.0019937}
}
Kostenlos starten →