|Table of Contents|

Citation:
　Lauri Vasankari,Adrian Borzyszkowski,Luca Zelioli,et al.Deep Mix: AI in Littoral Sonar Operations[J].Journal of Marine Science and Application,2026,(4):1018-1029.[doi:10.1007/s11804-025-00695-4]
Click and Copy

Deep Mix: AI in Littoral Sonar Operations

Info

Title:
Deep Mix: AI in Littoral Sonar Operations
Author(s):
Lauri Vasankari1; 2;  Adrian Borzyszkowski2;  Luca Zelioli2;  Jukka Heikkonen2
Affilations:
Author(s):
Lauri Vasankari1; 2;  Adrian Borzyszkowski2;  Luca Zelioli2;  Jukka Heikkonen2
1. Department of Military Technology, National Defence University, Kadettikouluntie 7, Helsinki 00860, Uusimaa, Finland;
2. Department of Computing, University of Turku, Vesilinnantie 3, Turku 20014, Varsinais-Suomi, Finland
Keywords:
Marine sensors; Mine countermeasures; Sonar and transducers; Convolutional neural network; Transfer learning; Vision transformer; Mixture of experts
分类号:
-
DOI:
10.1007/s11804-025-00695-4
Abstract:
The mine countermeasure operations consist of hunting i.e. searching for possible mines from the seabed and water column as well as demolition or disposal of found mines to enable safe passage of other naval units. Mine hunting is a task conducted by specialized naval units aiming to clear sea areas from estimated or realized naval mine threats. The search is usually initiated by sweeping broad areas with a side scan sonar to produce an initial picture of the seabed and localize possible threats for closer examination. The manual analysis process is challenging and time-consuming, especially in environments where seabeds are littered with mine-like objects such as rocks and other roughness. This paper researches the possibility of enhancing littoral mine hunting with object detection using a deep convolutional neural network and vision transformer methods separately and as a Mixture of Experts system to do the preliminary object detection and classification of the data multiples faster than a sole sonar operator. The results indicate that even a processed sonar image can be used to enhance mine hunting operations in a transparent, explainable manner but would benefit from enriching with additional data and signal analysis to ensure the detection of the most difficult targets.

References:

[1] Amit Y, Felzenszwalb P, Girshick R (2021) Object detection. Cham: Springer International Publishing, 875-883
[2] Armstrong A, Huff L, Glang G, Guenther G, Eisler T, Riley J, Perez S (1998) New technology for shallow water hydrographic surveys. International Hydrographie Review LXXV(2): 27-41
[3] Baird Z, Mcdonald MK, Rajan S, Lee SJ (2020) A CNN-lSTM network for augmenting target detection in real maritime wide area surveillance radar data. IEEE Access 8: 179281-179294
[4] Barngrover C, Kastner R, Belongie S (2015) Semisynthetic versus real-world sonar training data for the classification of mine-like objects. IEEE Journal of Oceanic Engineering 40(1): 48-56. https://doi.org/10.1109/JOE.2013.2291634
[5] Boser BE, Guyon IM, Vapnik VN (1992) A training algorithm for optimal margin classifiers. Proceedings of the Fifth Annual Workshop on Computational Learning Theory, COLT’92, New York, 144-152
[6] Breiman L (2001) Random forests. Machine Learning 45: 5-32. https://doi.org/10. 1023/A:1010933404324
[7] Brocki L, Chung NC (2023) Class-discriminative attention maps for vision transformers. ArXiv abs/2312.02364
[8] Broughton B (1998) The (R)evolution of mine countermeasures. U.S Naval Institute Proceedings 124(5): 1143
[9] Chandran V, Elgar S, Nguyen A (2002) Detection of mines in acoustic images using higher order spectral features. IEEE Journal of Oceanic Engineering 27(3): 610-618. https://doi.org/10.1109/joe.2002.1040943
[10] Chapple P (2008) Automated detection and classification in high-resolution sonar imagery for autonomous underwater vehicle operations. Technical Report DSTO-GD-0537, Maritime Operations Division, Defence Science and Technology Organisation
[11] Chapple PB (2009) Unsupervised detection of mine-like objects in seabed imagery from autonomous underwater vehicles. OCEANS 2009, 1-6
[12] Dosovitskiy A, Beyer L, Kolesnikov A, Weissenborn D, Zhai X, Unterthiner T, Dehghani M, Minderer M, Heigold G, Gelly S, Uszkoreit J, Houlsby N (2021) An image is worth 16×16 words: Transformers for image recognition at scale. arXiv preprint arXiv:2010.11929v2.arXiv:2010.11929v2 [cs.CV]
[13] Dwivedi R, Dave D, Naik H, Singhal S, Omer R, Patel P, Qian B, Wen Z, Shah T, Morgan G, Ranjan R (2023) Explainable AI (XAI): Core ideas, techniques, and solutions. ACM Comput. Surv. 55(9): 1-33. https://doi.org/10.1145/3561048
[14] Fei T, Kraus D, Zoubir AM (2015) Contributions to automatic target recognition systems for underwater mine classification. IEEE Transactions on Geoscience and Remote Sensing 53(1): 505-518. https://doi.org/10.1109/TGRS.2014.2324971
[15] Flemming BW (1976) Side-scan sonar: A practical guide. The International Hydrographic Review 53(1): 65-92
[16] Gebhardt D, Parikh K, Dzieciuch I, Walton M, Vo Hoang NA (2017) Hunting for naval mines with deep neural networks. OCEANS 2017-Anchorage, 1-5
[17] Gorman R, Sejnowski TJ (1988) Analysis of hidden units in a layered network trained to classify sonar targets. Neural Networks 1(1): 75-89. https://doi.org/10.1016/0893-6080(88)90023-8
[18] Gupta A, Ramanath R, Shi J, Keerthi SS (2021) Adam vs. SGD: Closing the generalization gap on image classification. OPT2021: 13th Annual Workshop on Optimization for Machine Learning, 1-7
[19] Hansen FS (2025) Russian military thinking about the Baltic Sea and the Arctic. DIIS Policy Brief, Danish Institute for International Studies (DIIS)
[20] Hayes MP, Gough PT (2009) Synthetic aperture sonar: A review of current status. IEEE Journal of Oceanic Engineering 34(3): 207-224. https://doi.org/10. 1109/JOE.2009.2020853
[21] Hibbert KR (2006) A need for systems architecture approach for next generation mine warfare capability. Master thesis, Naval Postgraduate School, Monterey, USA
[22] Ho?yń S (2021) A review of underwater mine detection and classification in sonar imagery. Electronics 10(23): 2943
[23] Hutchinson ML, Antono E, Gibbons BM, Paradiso S, Ling J, Meredig B (2017) Overcoming data scarcity with transfer learning. arXiv preprint arXiv:1711.05099
[24] Hyland JC, Dobeck GJ (1995) Sea mine detection and classification using sidelooking sonar. In A. C. Dubey, I. Cindrich, J. M. Ralston, and K. A. Rigano (Eds.), Detection Technologies for Mines and Minelike Targets, Volume 2496, 442-453
[25] Kaskela A (2017) Seabed landscapes of the Baltic Sea: Geological characterization of the seabed environment with spatial analysis techniques. PhD thesis, University of Helsinki, Helsinki, Finland
[26] Krizhevsky A, Sutskever I, Hinton GE (2012) Imagenet classification with deep convolutional neural networks. In F. Pereira, C. Burges, L. Bottou, and K. Weinberger (Eds.), Advances in Neural Information Processing Systems, Volume 25, 1097-1105
[27] Krizhevsky A, Sutskever I, Hinton GE (2017) Imagenet classification with deep convolutional neural networks. Commun. ACM 60(6): 84-90. https://doi.org/10.1145/3065386
[28] Kunde T, Schneider von Deimling J (2016) Hydroacoustic detection of dumped ammunition in the ocean with multibeam snippet backscatter analyses. A case study from the ’Kolberger Heide’ ammunition dump site (Baltic Sea, Germany). In EGU General Assembly Conference Abstracts, EGU General Assembly Conference Abstracts, EPSC2016-9168
[29] LeCun Y, Bengio Y (1998) Convolutional networks for images, speech, and time series. MIT Press, Cambridge, USA, 255-258
[30] Li W, Li C, Jiang H, Wang Y, Wu S, Wu Z (2024) High-resolution aerial image restoration with latent diffusion models. 2024 IEEE International Conference on Unmanned Systems (ICUS), 1874-1878
[31] Mansell T, Skinner D (1995) Application of Artificial Intelligence to decision making in Mine Countermeasures. Technical Report DSTO-TR-0279, Maritime Operations Division, Aeronautical and Maritime Research Laboratory
[32] Masoudnia S, Ebrahimpour R (2014) Mixture of experts: a literature survey. Artificial Intelligence Review 42: 275-293. https://doi.org/10.1007/s10462-012-9338-y
[33] Maurício J, Domingues I, Bernardino J (2023) Comparing vision transformers and convolutional neural networks for image classification: A literature review. Applied Sciences 13(9): 5521. https://doi.org/10.3390/app13095521
[34] McKay J, Gerg I, Monga V, Raj R (2017) What’s mine is yours: Pretrained CNNs for limited training sonar ATR. OCEANS 2017-Anchorage, Anchorage, USA, 1-7
[35] Melia T (1991) Damn the Torpedoes: A Short History of U.S. Naval Mine Countermeasures, 1777-1991. Contributions to Naval History, Naval Historical Center
[36] Munteanu D, Moina D, Zamfir CG, Petrea M, Cristea DS, Munteanu N (2022) Sea mine detection framework using YOLO, SSD and EfficientDet deep learning models. Sensors 22(23): 9536. https://doi.org/10.3390/s22239536
[37] Peine H, Brecht D (2005) Detection of objects buried in the seafloor. Europe Oceans 2005, Volume 1, 177-181
[38] Perez L, Wang J (2017) The effectiveness of data augmentation in image classification using deep learning. arXiv preprint arXiv:1712.04621
[39] Petillot Y, Reed S, Coiras E, Bell J (2006) A framework for evaluating underwater mine detection and classification algorithms using augmented reality. Technical report, Heriot-Watt University
[40] Prusty S, Patnaik S, Dash SK (2022) SKCV: Stratified K-fold cross-validation on ML classifiers for predicting cervical cancer. Frontiers in Nanotechnology 4: 972421
[41] Redmon J, Divvala S, Girshick R, Farhadi A (2016) You only look once: Unified, real-time object detection. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 779-788
[42] Schwarz M (2014) Future mine countermeasures: No easy solutions. Naval War College Review 67(3): 123-141
[43] Selvaraju RR, Cogswell M, Das A, Vedantam R, Parikh D, Batra D (2017) Grad-CAM: Visual explanations from deep networks via gradient-based localization. Proceedings of the IEEE International Conference on Computer Vision, 618-626
[44] Simonyan K, Zisserman A (2015) Very deep convolutional networks for large-scale image recognition. 3rd International Conference on Learning Representations, 1-14
[45] Sun Y, Zheng H, Zhang G, Ren J, Xu H, Xu C (2022) DP-ViT: A dual-path vision transformer for real-time sonar target detection. Remote Sensing 14(22): 5807. https://doi.org/10.3390/rs14225807
[46] Thanh Le H, Phung SL, Chapple PB, Bouzerdoum A, Ritz CH, Tran LC (2020) Deep gabor neural network for automatic detection of mine-like objects in sonar imagery. IEEE Access 8: 94126-94139. https://doi.org/10.1109/ACCESS. 2020.2995390
[47] Tueller P, Kastner R, Diamant R (2018) A comparison of feature detectors for underwater sonar imagery. OCEANS 2018 MTS/IEEE Charleston, 1-6
[48] Vaswani A, Shazeer N, Parmar N, Uszkoreit J, Jones L, Gomez AN, Kaiser Lu, Polosukhin I (2017) Attention is all you need. In I. Guyon, U. V. Luxburg, S. Bengio, H.Wallach, R. Fergus, S. Vishwanathan, and R. Garnett (Eds.), Advances in Neural Information Processing Systems, Curran Associates, Inc., Volume 30
[49] Williams DP (2016) Underwater target classification in synthetic aperture sonar imagery using deep convolutional neural networks. 2016 23rd International Conference on Pattern Recognition (ICPR), 2497-2502
[50] Wu M, Wang Q, Rigall E, Li K, Zhu W, Bo H, Yan T (2019) ECNet: Efficient convolutional networks for side scan sonar image segmentation. Sensors (Basel, Switzerland) 19(9): 48-56. https://doi.org/10.3390/s19092009
[51] Yao D, Azimi-Sadjadi M, Jamshidi A, Dobeck G (2002) A study of effects of sonar bandwidth for underwater target classification. IEEE Journal of Oceanic Engineering 27(3): 619-627. https://doi.org/10.1109/JOE.2002.1040944
[52] Yuksel SE, Wilson JN, Gader PD (2012) Twenty years of mixture of experts. IEEE Transactions on Neural Networks and Learning Systems 23(8): 1177-1193. https://doi.org/10.1109/TNNLS.2012. 2200299
[53] Zhang Z, Sabuncu MR (2018) Generalized cross entropy loss for training deep neural networks with noisy labels. Proceedings of the 32nd International Conference on Neural Information Processing Systems, NIPS’18, New York, USA, 8792-8802
[54] Zhao ZQ, Zheng P, Xu ST, Wu X (2019) Object detection with deep learning: A review. IEEE Transactions on Neural Networks and Learning Systems 30(11): 3212-3232. https://doi.org/10.1109/TNNLS.2018.2876865

Memo

Memo:
Received date:2025-2-15;Accepted date:2025-3-17。<br>Foundation item:Funding This research was funded by MATINE (Finnish National Defence Scientific Advisory Board), Grant Number VN/15392/2022-SAAP/5 in collaboration with the University of Turku. Authors from the University of Turku have also received a research grant from the Department of Computing.<br>Corresponding author:Lauri Vasankari，E-mail:lauri.vasankari@gmail.com
Last Update: 2026-10-16