Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Video

Updated: May 2, 2026

Long-term Behavioral Tracking of Freely Swimming Weakly Electric Fish
10:56

Long-term Behavioral Tracking of Freely Swimming Weakly Electric Fish

Published on: March 6, 2014

12.2K

A few-shot learning method for underwater acoustic target recognition based on generative data augmentationa).

Wei Huang1, Qirui Zhang1, Bowen Zhao1

  • 1Faculty of Information Science and Engineering, Ocean University of China, Qingdao, Shandong 266100, China.

The Journal of the Acoustical Society of America
|May 1, 2026
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

TRUST: A toolkit for TEE-assisted secure outsourced computation over integers.

Fundamental research·2026
Same author

Uncovering the Key Circuit FOSL2/FOS/EGR3/EGR1, Contributing to the Hyperexcitability of Excitatory Neurons in the Epileptic Temporal Cortex and Hippocampus.

International journal of molecular sciences·2026
Same author

MixNet: A scale-adaptive method for multivariate time series forecasting.

PloS one·2026
Same author

How Does Right Atrial Function Differ? A Speckle-Tracking Echocardiographic Comparison of Fetuses With Left Ventricular Outflow Tract Obstruction Versus Those With Significant Tricuspid Regurgitation.

Echocardiography (Mount Kisco, N.Y.)·2026
Same author

In-situ chlorine-induced dynamic diatomic reconstruction for three electron transfer mineralization of high-salinity organic wastewater.

Water research·2026
Same author

Prognostic implications of dominant lesion misidentification in multiple primary lung cancer.

Translational lung cancer research·2026

This study introduces a new deep learning framework to improve underwater acoustic target recognition (UATR) despite limited data. The method enhances feature discrimination for more robust UATR systems.

Area of Science:

  • Marine acoustics
  • Deep learning
  • Signal processing

Background:

  • Underwater acoustic target recognition (UATR) is crucial for marine protection and national security.
  • Deep learning methods for UATR face challenges due to limited reference samples and environmental interference.
  • Existing methods struggle with incomplete intra-class data distributions and unstable classification boundaries.

Purpose of the Study:

  • To propose a novel generative discriminative collaborative framework for robust UATR.
  • To address the challenges of data scarcity and environmental interference in UATR.
  • To enhance the continuity, integrity, and discriminative power of feature-level representations.

Main Methods:

  • A variational auto-encoder boosted learning framework based on latent space completion.

More Related Videos

Development of New Methods for Quantifying Fish Density Using Underwater Stereo-video Tools
09:32

Development of New Methods for Quantifying Fish Density Using Underwater Stereo-video Tools

Published on: November 20, 2017

8.8K

Related Experiment Videos

Last Updated: May 2, 2026

Long-term Behavioral Tracking of Freely Swimming Weakly Electric Fish
10:56

Long-term Behavioral Tracking of Freely Swimming Weakly Electric Fish

Published on: March 6, 2014

12.2K
Development of New Methods for Quantifying Fish Density Using Underwater Stereo-video Tools
09:32

Development of New Methods for Quantifying Fish Density Using Underwater Stereo-video Tools

Published on: November 20, 2017

8.8K
  • A structure-preserving generative reconstruction mechanism to supplement datasets implicitly.
  • A three-stage pipeline: unified preprocessing, multi-scale latent space reconstruction, and acoustic identification.
  • A staged modeling workflow for synergistic data purification, latent space completion, and discriminative optimization.

    • Main Results:

    • The framework reconstructs more continuous and discriminative intra-class distributions at the feature level.
    • Implicit data supplementation enhances the completeness of intra-class manifolds.
    • The proposed method establishes a robust recognition paradigm for few-shot learning scenarios in UATR.
    • Improved feature-level discrimination and stability of classification boundaries were achieved.

    Conclusions:

    • The generative discriminative collaborative framework offers a robust solution for UATR with limited data.
    • Latent space completion and generative reconstruction are effective in overcoming data scarcity.
    • The staged pipeline optimizes individual objectives while maintaining overall synergy for enhanced UATR performance.