Based on One-dimensional Attention Mechanism Convolutional Neural Network for Underwater Acoustic Target Recognition

ZHANG Yufei; ZHAO Mei; HU Changqing; GUO Zheng

doi:10.11993/j.issn.2096-3920.2025-0053

Article Contents

Article Navigation > Journal of Unmanned Undersea Systems > 2025 > Accepted Manuscript

ZHANG Yufei, ZHAO Mei, HU Changqing, GUO Zheng. Based on One-dimensional Attention Mechanism Convolutional Neural Network for Underwater Acoustic Target Recognition[J]. Journal of Unmanned Undersea Systems. doi: 10.11993/j.issn.2096-3920.2025-0053

Citation:

ZHANG Yufei, ZHAO Mei, HU Changqing, GUO Zheng. Based on One-dimensional Attention Mechanism Convolutional Neural Network for Underwater Acoustic Target Recognition[J]. Journal of Unmanned Undersea Systems. doi: 10.11993/j.issn.2096-3920.2025-0053

Citation:

PDF( 1531 KB)

Based on One-dimensional Attention Mechanism Convolutional Neural Network for Underwater Acoustic Target Recognition

doi: 10.11993/j.issn.2096-3920.2025-0053

ZHANG Yufei^{1, 2
,},
ZHAO Mei^1
,,
HU Changqing^1
,,
GUO Zheng^1
,

1.
Shanghai Acoustics Laboratory, Chinese Academy of Sciences, Shanghai 201815, China
2.
University of Chinese Academy of Sciences, Beijing 100049, China

Received Date: 2025-04-07
Accepted Date: 2025-05-20
Rev Recd Date: 2025-05-18

Available Online: 2025-09-12

Abstract

Abstract

To address the issues of complex network parameters and high computational costs in deep learning-based underwater acoustic target recognition models, this study proposes a lightweight one-dimensional convolutional neural network with an attention mechanism for underwater acoustic target recognition.First, during the feature extraction stage, spectral, Mel-spectrogram, chroma, spectral contrast, and tonal features are selected and reconstructed into a fused one-dimensional hybrid feature. Next, the hybrid feature is processed by a multi-scale residual convolution (MRC) module to enhance feature representation across different scales. Simultaneously, a Convolutional Block Attention Module (CBAM) is introduced to adaptively adjust feature importance through channel and spatial attention mechanisms, improving the model's focus on critical regions.Experimental results show that the proposed model achieves an average recognition accuracy of 98.58% on the ShipsEar dataset, demonstrating excellent classification performance. Compared to existing models, this model significantly reduces computational complexity. Further validation on real-world data from the East China Sea confirms its effectiveness.
- underwater acoustic target recognition,
- multi-feature fusion,
- convolutional block attention module,
- 1D neural network

FullText(HTML)

References(24)

References

[1]	徐及, 黄兆琼, 李琛, 颜永红. 深度学习在水下目标被动识别中的应用进展[J]. 信号处理, 2019, 35(9): 1460-1475. XU J, HUANG Z Q, LI C, et al. Advances in applications of deep learning in passive underwater target recognition[J]. Signal Processing, 2019, 35(9): 1460-1475.
[2]	蔡悦斌, 张明之, 史习智, 等. 舰船噪声波形结构特征提取及分类研究[J]. 电子学报, 1999(06): 129-130. doi: 10.3321/j.issn:0372-2112.1999.06.030 CAI Y B, ZHANG M Z, SHI X Z, et al. Research on extraction and classification of structural features of ship-radiated noise waveform[J]. Acta Electronica Sinica, 1999(06): 129-130. doi: 10.3321/j.issn:0372-2112.1999.06.030
[3]	葛轶洲, 姚泽, 张歆, 等. 水声目标的MFCC特征提取与分类识别[J]. 计算机仿真, 2024, 41(02): 13-16+33. doi: 10.3969/j.issn.1006-9348.2024.02.003 GE Y Z, YAO Z, ZHANG X, et al. MFCC feature extraction and classification recognition of underwater acoustic targets[J]. Computer Simulation, 2024, 41(02): 13-16+33. doi: 10.3969/j.issn.1006-9348.2024.02.003
[4]	谢东日. 水声目标非线性特征提取研究[D]. 厦门: 厦门大学, 2020.
[5]	倪俊帅, 赵梅, 胡长青. 基于深度学习的舰船辐射噪声多特征融合分类[J]. 声学技术, 2020, 39(03): 366-371. NI J S, ZHAO M, HU C Q. Multi-feature fusion classification of ship-radiated noise based on deep learning[J]. Technical Acoustics, 2020, 39(03): 366-371.
[6]	梁喆, 侯朋, 夏春艳, 等. 融合时频域特征的舰船识别方法及实验研究[J]. 声学技术, 2021, 40(05): 607-613. LIANG Z, HOU P, XIA C Y, et al. Ship recognition method and experimental study based on fusion of time-frequency domain features[J]. Technical Acoustics, 2021, 40(05): 607-613.
[7]	刘承伟, 洪峰, 冯海泓, 胡梦璐. 结合多尺度卷积网络和双端注意力机制的水声目标识别[J]. 声学技术, 2023, 42(02): 161-167. LIU C W, HONG F, FENG H H, HU M L. Underwater acoustic target recognition combining multi-scale convolutional networks and dual attention mechanism[J]. Technical Acoustics, 2023, 42(02): 161-167.
[8]	杨基睿, 鄢社锋, 曾迪, 等. 改进通道注意力机制的时域水声信号识别网络[J]. 信号处理, 2023, 39(6): 1025-1035. YANG J R, YAN S F, ZENG D, et al. Time-domain underwater acoustic signal recognition network with improved channel attention mechanism[J]. Signal Processing, 2023, 39(6): 1025-1035.
[9]	LIU Y X, ZHANG B Q, KONG F T, et al. Underwater acoustic classification using wavelet scattering transform and convolutional neural network with limited dataset[J]. Applied Acoustics, 2025, 232: 110564. doi: 10.1016/j.apacoust.2025.110564
[10]	LYU C G, HU X Y, NIU Z H, et al. A light-weight neural network for marine acoustic signal recognition suitable for fiber-optic hydrophones, Expert Systems with Applications, 2024, 235: 121235.
[11]	杨路飞, 章新华, 吴秉坤, 等. 基于MFCC特征的被动水声目标深度学习分类方法[J]. 舰船科学技术, 2020, 42(19): 129-133. doi: 10.3404/j.issn.1672-7649.2020.10.025 YANG L F, ZHANG X H, WU B K, et al. Passive underwater acoustic target classification method based on MFCC features and deep learning[J]. Ship Science and Technology, 2020, 42(19): 129-133. doi: 10.3404/j.issn.1672-7649.2020.10.025
[12]	ZHANG S K, TIAN D Y, WANG C, et al. Intelligent recognition of underwater acoustic target noise on underwater glider platform[C]//2018 Chinese Automation Congress(CAC). Piscataway, NJ: IEEE, 2018: 4189-4193.
[13]	任晨曦, 王黎明, 韩星程, 等. 基于联合神经网络的水声目标识别方法[J]. 舰船科学技术, 2022, 44(1): 136-141. doi: 10.3404/j.issn.1672-7649.2022.01.026 REN C X, WANG L M, HAN X C, et al. Underwater acoustic target recognition method based on joint neural network[J]. Ship Science and Technology, 2022, 44(1): 136-141. doi: 10.3404/j.issn.1672-7649.2022.01.026
[14]	YANG K, WANG B, FANG Z, et al. An end-to-end underwater acoustic target recognition model based on one-dimensional convolution and transformer[J]. Journal of Marine Science and Engineering. 2024; 12(10): 1793.
[15]	LIU D Y, YANG H Y, HOU W M, et al. A novel underwater acoustic target recognition method based on MFCC and RACNN[J]. Sensors, 2024, 24(1): 273. doi: 10.3390/s24010273
[16]	Yücesoy, E. Gender recognition based on the stacking of different acoustic features[J]. Applied Sciences, 2024, 14(15): 6564. doi: 10.3390/app14156564
[17]	ZHANG S, GAO Y, CAI, J, et al. A novel bird sound recognition method based on multi-feature fusion and a transformer encoder[J]. Sensors, 2023, 23(19): 8099. doi: 10.3390/s23198099
[18]	WOO S, PARK J, LEE J Y, et al. CBAM: Convolutional block attention module[J]. European Conference on Computer Vision. 2018, 1121, 1: 3-19.
[19]	DAVID S D, TORRES G S , ANTONIO C L, et al. ShipsEar: An underwater vessel noise database[J]. Applied Acoustics, 2016, 113: 64-69.
[20]	YANG S, JIN A, ZENG X, et al. Underwater acoustic target recognition based on sub-band concatenated Mel spectrogram and multidomain attention mechanism[J]. Engineering Applications of Artificial Intelligence, 2024, 133: 107983. doi: 10.1016/j.engappai.2024.107983
[21]	NI J, Ji F, LU S, et al. Underwater target recognition method based on singular spectrum analysis and channel attention convolutional neural network[J]. Sensors, 2025, 25: 2573. doi: 10.3390/s25082573
[22]	WANG B, ZHANG W, ZHU Y, et al. An underwater acoustic target recognition method based on AMNet[J]. IEEE Geoscience and Remote Sensing Letters, 2023, 20: 1-5.
[23]	WANG J Y, QIAN P, CHEN Y X, et al. Adaptive underwater acoustic target recognition based on multi-scale residual and attention mechanism[J]. Digital Signal Processing, 2025, 163: 105193. doi: 10.1016/j.dsp.2025.105193
[24]	NI J, Ji F, LU S, et al. An Auditory Convolutional Neural Network for Underwater Acoustic Target Timbre Feature Extraction and Recognition[J]. Remote Sens. 2024, 16(16), 3074.