Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Adversarial multi-task underwater acoustic target recognition: towards robustness against various influential factors

Yuan Xie, Ji Xu, Jiawei Ren, Junfeng Li

Abstract

Underwater acoustic target recognition based on passive sonar faces numerous challenges in practical maritime applications. One of the main challenges lies in the susceptibility of signal characteristics to diverse environmental conditions and data acquisition configurations, which can lead to instability in recognition systems. While significant efforts have been dedicated to addressing these influential factors in other domains of underwater acoustics, they are often neglected in the field of underwater acoustic target recognition. To overcome this limitation, this study designs auxiliary tasks that model influential factors (e.g., source range, water column depth, or wind speed) based on available annotations and adopts a multi-task framework to connect these factors to the recognition task. Furthermore, we integrate an adversarial learning mechanism into the multi-task framework to prompt the model to extract representations that are robust against influential factors. Through extensive experiments and analyses on the ShipsEar dataset, our proposed adversarial multi-task model demonstrates its capacity to effectively model the influential factors and achieve state-of-the-art performance on the 12-class recognition task.

Adversarial multi-task underwater acoustic target recognition: towards robustness against various influential factors

Abstract

Underwater acoustic target recognition based on passive sonar faces numerous challenges in practical maritime applications. One of the main challenges lies in the susceptibility of signal characteristics to diverse environmental conditions and data acquisition configurations, which can lead to instability in recognition systems. While significant efforts have been dedicated to addressing these influential factors in other domains of underwater acoustics, they are often neglected in the field of underwater acoustic target recognition. To overcome this limitation, this study designs auxiliary tasks that model influential factors (e.g., source range, water column depth, or wind speed) based on available annotations and adopts a multi-task framework to connect these factors to the recognition task. Furthermore, we integrate an adversarial learning mechanism into the multi-task framework to prompt the model to extract representations that are robust against influential factors. Through extensive experiments and analyses on the ShipsEar dataset, our proposed adversarial multi-task model demonstrates its capacity to effectively model the influential factors and achieve state-of-the-art performance on the 12-class recognition task.

Paper Structure

This paper contains 19 sections, 6 equations, 6 figures, 7 tables.

Table of Contents

  1. Introduction
  2. Related Works
  3. Methods
  4. Pre-Processing and Feature Extraction
  5. Design of Auxiliary Tasks
  6. Detailed Process of Model Training and Testing
  7. Model Architecture
  8. Optimization Techniques for AMT
  9. Experimental Setup
  10. Dataset
  11. Data division
  12. Parameters Setup
  13. Results and Analyses
  14. Preliminary Experiments
  15. Comparative Experiments
  16. ...and 4 more sections

Figures (6)

  • Figure 1: A comparison of signal waveforms and spectrograms belonging to the same target (a passenger ship "Minho Uno") under different influential factors. The samples are drawn from the ShipsEar dataset.
  • Figure 2: The detailed process of data acquisition and feature extraction. Parameter setups or feature dimensions are displayed in red characters.
  • Figure 3: The distribution of different categories of data in the ShipsEar dataset under various influential factors: (a) source ranges; (b) water column depths; (c) wind speeds.
  • Figure 4: The detailed process of the training and test stage of our proposed AMT.
  • Figure 5: Preliminary experiments on selecting the optimal filter cutoff frequency, backbone model architecture, and acoustic feature.
  • ...and 1 more figures