EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Deep Learning Based Underwater Acoustic Target Recognition: Introduce a Recent Temporal 2D Modeling Method

Jun Tang iD, Wenbo Gao iD, Enxue Ma, Xinmiao Sun, Jinying Ma

Sensors · 02.03.2024 · 24 (5) · 1633

DOI: 10.3390/s24051633 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Sensors, online erschienen am 02.03.2024. Verfasst von Jun Tang und 4 weiteren Autor:innen. Beteiligte Einrichtung: School of Civil Engineering, Tianjin University, Tianjin 300072, China. Die Arbeit wurde bisher 14-mal zitiert (Crossref) und nennt 38 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: the National Key R&D Program of China, National Key R&D Program of China.

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In recent years, the application of deep learning models for underwater target recognition has become a popular trend. Most of these are pure 1D models used for processing time-domain signals or pure 2D models used for processing time-frequency spectra. In this paper, a recent temporal 2D modeling method is introduced into the construction of ship radiation noise classification models, combining 1D and 2D. This method is based on the periodic characteristics of time-domain signals, shaping them into 2D signals and discovering long-term correlations between sampling points through 2D convolution to compensate for the limitations of 1D convolution. Integrating this method with the current state-of-the-art model structure and using samples from the Deepship database for network training and testing, it was found that this method could further improve the accuracy (0.9%) and reduce the parameter count (30%), providing a new option for model construction and optimization. Meanwhile, the effectiveness of training models using time-domain signals or time-frequency representations has been compared, finding that the model based on time-domain signals is more sensitive and has a smaller storage footprint (reduced to 30%), whereas the model based on time-frequency representation can achieve higher accuracy (1–2%).

Schlagwörter

Stichwörter der Autor:innen

deep learningshort-time Fourier transformtemporal 2D modelingunderwater acoustic target recognition

Autor:innen und Institutionen

  • School of Civil Engineering, Tianjin University, Tianjin 300072, China.
  • School of Electrical and Information Engineering, Tianjin University, Tianjin 300072, China.
  • School of Electronic Engineering, Tianjin University of Technology and Education, Tianjin 300222, China.

ORCID-Profile: Jun Tang, Wenbo Gao

Förderung

  • the National Key R&D Program of China
  • National Key R&D Program of China

Zitieren

APA

Tang, J., Gao, W., Ma, E., Sun, X., & Ma, J. (2024). Deep Learning Based Underwater Acoustic Target Recognition: Introduce a Recent Temporal 2D Modeling Method. Sensors, 24(5), 1633. https://doi.org/10.3390/s24051633

Vancouver

Tang J, Gao W, Ma E, Sun X, Ma J. Deep Learning Based Underwater Acoustic Target Recognition: Introduce a Recent Temporal 2D Modeling Method. Sensors. 2024;24(5):1633. doi:10.3390/s24051633

BibTeX
@article{Tang2024,
  author = {Tang, Jun and Gao, Wenbo and Ma, Enxue and Sun, Xinmiao and Ma, Jinying},
  title = {Deep Learning Based Underwater Acoustic Target Recognition: Introduce a Recent Temporal 2D Modeling Method},
  journal = {Sensors},
  year = {2024},
  volume = {24},
  number = {5},
  pages = {1633},
  doi = {10.3390/s24051633}
}
Kostenlos starten →