A self-supervised dual-channel self-attention acoustic encoder for underwater acoustic target recognition
Publication Type
Journal Article
Publication Date
5-2024
Abstract
Due to the considerable noise and high annotation cost of passive sonar acquisition data challenges, underwater acoustic target recognition (UATR) tasks require novel solutions. In this paper, we propose a self-supervised dual-channel self-attention acoustic encoder (DSAE) for UATR tasks. First, to address the annotation challenge, we design a dual-channel self-attention acoustic encoder to unify features into self-supervised learning. Meanwhile, we propose a dynamic positive sample memory module (DMM) to enable the training samples comprehensive and balanced in self-supervised learning. Second, to address the noise challenge, we utilize a time–frequency mask to obtain space–time enhanced features. Based on our experimental results, DSAE improves the recognition accuracy and anti-noise robustness compared with other advanced acoustic learning methods. The results demonstrate that DSAE has the potential to offer significant value as a tool for UATR tasks.
Keywords
Dual-channel mechanism, Local self-attention, Self-supervised learning, Underwater acoustic target recognition
Discipline
Artificial Intelligence and Robotics | Oceanography and Atmospheric Sciences and Meteorology
Publication
Ocean Engineering
Volume
299
ISSN
0029-8018
Identifier
10.1016/j.oceaneng.2024.117305
Publisher
Elsevier
Citation
WANG, Xingmei; WU, Peiran; LI, Boquan; ZHAN, Ge; LIU, Jinghan; and LIU, Zijian.
A self-supervised dual-channel self-attention acoustic encoder for underwater acoustic target recognition. (2024). Ocean Engineering. 299,.
Available at: https://ink.library.smu.edu.sg/sis_research/8738
Additional URL
https://doi.org/10.1016/j.oceaneng.2024.117305