Deep learning based target recognition system under limited and imbalanced dataset for monostatic active sonar using range and bearing features

Yena You; Wonnyoung Lee; Seokjin Lee

doi:10.7776/ASK.2025.44.6.568

All Issue

2025 Vol.44, Issue 6 Preview Page Next Page

Research Article

Deep learning based target recognition system under limited and imbalanced dataset for monostatic active sonar using range and bearing features 소량 불균형 데이터셋의 거리 및 방위 특징을 활용한 딥러닝 모델 기반 단상태 능동 소나 표적 식별 시스템 연구

30 November 2025. pp. 568-579

PDF XML

Abstract

This paper proposes a system for a automatic active sonar target classification system that utilizes target detection images in the bearing-range domain as input feature. These images are obtained by applying conventional active sonar signal processing techniquis such as beam-forming and matched filtering to the received signals. Deep Convolutional Neural Networks (DCNN) and Convolutional Recurrent Neural Network (CRNN) were employed as classifiers, and their target classification performance was compared with respect to the proposed input features. Two simulation datasets were generated using a simulator under a monostatic active sonar scenario, each assuming different ocean environments, and were used for model training. The performance of the proposed system was evaluated on validation datasets using accuracy as well as target and non-target F1-score as metrics. The experimental results demonstrate that the proposed features, which incorporate both bearing and long-range detection information, can be effectively exploited by neural network architectures regardless of the transmitted pulse type. Also, the meaningful target classification performance can be achieved even with a limited amount of imbalanced training data without the application of additional data augmentation.

Keywords

Active sonar

Automatic target recognition

Beam-forming

Matched filter

Neural network

본 논문은 능동 소나 시스템에서 원통형 배열 센서에 수신된 신호에 대해 빔 형성과 정합 필터와 같은 고전적인 능동 소나 신호처리 기법을 적용하여 얻은 거리-방위 영역의 표적 탐지 이미지를 특징으로 활용한 능동 소나 자동 표적 식별기 시스템을 제안한다. 이때 심층 합성곱 신경망(Deep Convolutional Neural Network, DCNN)과 합성곱 순환 신경망(Convolutional Recurrent Neural Network, CRNN)을 표적 식별기에 적용하였으며 제안하는 입력 데이터 특징에 대한 두 모델의 표적 식별 성능을 비교했다. 단상태 능동 소나 시스템을 가정하는 시뮬레이터 시나리오를 통해 얻은 두 시뮬레이션 데이터 세트를 사용하여 모델 학습을 수행하였고 각각의 데이터 세트는 서로 다른 해양 환경을 가정한다. 제안하는 시스템의 표적 식별 성능은 독립된 검증용 데이터 세트를 통해 얻은 정확도와 표적 및 비표적 F1 점수를 지표로 평가한다. 본 논문을 통해 송신 펄스의 유형과 무관하게 방위 및 장거리 탐지 정보를 포함하는 표적 탐지 이미지 특징을 신경망 구조에 활용할 수 있음을 확인하였으며 추가적인 데이터 증강 기법이 적용되지 않은 소량의 불균형 데이터만으로 유의미한 표적 식별 성능을 달성할 수 있음을 보였다.

키워드

능동 소나

자동 표적 식별

빔 형성

정합 필터

신경망 구조

References

R. J. Ulrich, Principles of Underwater Sound for Engineers (McGraw-Hill Book, New York, 1967), pp. 1-342.

E. Miasnikov, “Can Russian strategic submarines survive at sea? The fundamental limits of passive acoustics,” Sci. Glob. Secur. 4, 213-251 (1994).

10.1080/08929889408426401

B. N. Gross, Input of factor graphs into the detection, classification, and localization chain and continuous active sonar in undersea vehicles, (M.S. thesis, Verginia Tech, 2015).

M. Najibzadeh, A. Mahmoodzadeh, and M. Khishe, “Active sonar image classification using deep convolutional neural network evolved by robust comprehensive grey wolf optimizer,” Neural Process. Lett. 55, 8689-8712 (2023).

10.1007/s11063-023-11173-9

I. Seo, S. Kim, Y. Ryu, J. Park, and S. Han, “Underwater moving target classification using multilayer processing of active sonar system,” Appl. Sci. 9, 4617 (2019).

10.3390/app9214617

Y. You, W. Lee, and S. Lee, “A study on the weakly-supervised deep learning algorithm for active sonar target recognition based on pseudo labeling using convolutional recurrent neural network model” (in Korean), J. Acoust. Soc. Kr. 43, 502-510 (2024).

D.-H. Lee, “Pseudo-label: The simple and efficient semi-supervised learning method for deep neural networks,” Proc. ICML, 896 (2013).

A. D. Waite, Sonar for Practisting Engineers (Wiley, Chichester, 2002), pp. 192-197.

I. Goodfellow, Y. Bengio, and A. Courville, DeepLearning (MIT press, Cambridge, 2016), pp. 1-800.

O. Russakovsky, J. Deng, H. Su, J. Krause, S. Satheesh, S. Ma, Z. Huang, A. Karpathy, A. Khosla, M. Bernstein, A. C. Berg, and L. Fei-Fei, “Imagenet large scale visual recognition challenge,” IJCV, 15, 211-252 (2015).

10.1007/s11263-015-0816-y

B. Shi, X. Bai, and C. Yao, “An end-to-end trainable neural network for image-based sequence recognition and its application to scene text recognition,” IEEE Trans. Pattern Anal. Mach. Intell. 39, 2298-2304 (2016).

10.1109/TPAMI.2016.2646371

D. A. Abraham, Underwater Acoustic Signal Processing: Modeling, Detection, and Estimation (Springer, Berlin, 2019), Chapter.2.

10.1007/978-3-319-92983-5

N. Turpault, R. Serizel, A. Shah, and J. Salamon, “Sound event detection in domestic environments with weakly labeled data and soundscape synthesis,” Proc. DCASE. 1-5 (2019).

10.33682/006b-jx26

L. Delphin-Poulat, and C. Plapous, “Mean teacher with data augmentation for dcase 2019 task 4,” Orange Labs Lannion, Tech. Rep., 2019.

Information

Publisher :The Acoustical Society of Korea
Publisher(Ko) :한국음향학회
Journal Title :The Journal of the Acoustical Society of Korea
Journal Title(Ko) :한국음향학회지
Volume : 44
No :6
Pages :568-579
Received Date : 2025-09-16
Revised Date : 2025-10-28
Accepted Date : 2025-11-17
DOI :https://doi.org/10.7776/ASK.2025.44.6.568

The Journal of the Acoustical Society of KoreaISSN:1225-4428(Print) 2287-3775(Online)한국음향학회

All Issue