According to US National Oceanic and Atmospheric Administration (NOAA), we have only explored five percent of our world ocean. Among the ocean exploration tasks, underwater object detection is a multidisciplinary challenge that extends beyond engineering into fields such as oceanography, marine biology, and environmental science. Scientifically, it involves understanding the complex interactions between sound waves, light, and water, as well as the navigation dynamics within marine environments. The underwater environment presents unique physical challenges for object detection due to factors like light attenuation, turbidity, and the variability of acoustic propagation. This paper introduces a cost-effective autonomous underwater prototype for real-time detection, and localization of small underwater objects (e.g., archaeological artifacts, parts of infrastructures, wreckage debris, but also life forms like fishes, corals etc.) in shallow waters. The system combines object identification with autonomous navigation capabilities. It consists of an autonomous underwater vehicle equipped with sensors, cameras, and localization tools, as well as a ground control station for monitoring and intervention. In particular, we focus on a case study about detecting and reporting the locations of unexploded ordnance materials, contributing to the monitoring of underwater hazards in conflict-affected regions. Key contributions include the integration of a cost-effective autonomous remotely operated vehicle with sensors and software for real-time detection and localization of small underwater objects, as well as an annotated dataset of UXO images, usable as a benchmark.
ChenJErMJ. Dynamic YOLO for small underwater object detection. Artif Intell Rev2024; 57: 1–23.
5.
LinTYMaireMBelongieS, et al.Microsoft COCO: common objects in context. In: Fleet D, Pajdla T, Schiele B and Tuytelaars T (eds) Computer vision – ECCV 2014. Cham: Springer International Publishing, 2014. pp.740–55.
6.
Guerrero-RodriguezBGarcia-RodriguezJSalvadorJ, et al.Improving landslide prediction by computer vision and deep learning. Integr Comput Aided Eng2023; 31: 77–94.
7.
LiDLiuMYangL, et al.A non-contact identification method of overweight vehicles based on computer vision and deep learning. Comput Aid Civil Infrast Eng2024; 39: 3452–3476.
8.
PanXYangTYXiaoY, et al.Vision-based real-time structural vibration measurement through deep-learning-based detection and tracking methods. Eng Struct2023; 281: 115676.
9.
JiangTFrøsethGTRønnquistA, et al.A visual inspection and diagnosis system for bridge rivets based on a convolutional neural network. Comput Aid Civil Infrast Eng2024; 39: 3786–3804.
10.
JiangTLiLSamaliB, et al.Lightweight object detection network for multi-damage recognition of concrete bridges in complex environments. Comput Aid Civil Infrast Eng2024; 39: 3646–3665.
11.
ShenJJiaoLZhangC, et al.Monocular 3D object detection for construction scene analysis. Comput Aid Civil Infrast Eng2024; 39: 1370–1389.
12.
UrdialesJMartínDArmingolJM. An improved deep learning architecture for multi-object tracking systems. Integr Comput Aided Eng2023; 30: 121–134.
13.
ChaverotMCarréMJourlinM, et al.Improvement of small objects detection in thermal images. Integr Comput Aided Eng2023; 30: 1–15.
14.
SummeyDCMcCormickJFCarrollPJ. Mobile underwater debris survey system (MUDSS). In: Oceans ’99. MTS/IEEE. riding the crest into the 21st century. Conference and exhibition. Conference Proceedings (IEEE Cat. No.99CH37008). vol. 1; 1999. pp.363–72.
15.
BrownDCJohnsonSFGergID, et al.Simulation and testing results for a sub-bottom imaging sonar. Proc Meet Acoust2019; 36: 070001.
16.
BucaroJHoustonBSanigaM, et al.Wide area detection and identification of underwater uxo using structural acoustic sensors. In: Naval research laboratory. vol. 12, 2007, pp.1–86. Washington DC.
17.
BucaroJAWatersZJHoustonBH, et al.Acoustic identification of buried underwater unexploded ordnance using a numerically trained classifier (L). J Acoust Soc Am2012; 132: 3614–3617.
18.
SERDP. Workshop on acoustic detection and classification of uxo in the underwater environment. Final Report. U.S. Department of Defense, 2013.
19.
SERDP. Workshop on acoustic detection and classification of munitions in the underwater environment. Final Report. U.S. Department of Defense, 2018.
20.
BeaujeanPPJBrissonLNNegahdaripourS. High-resolution imaging sonar and video technologies for detection and classification of underwater munitions. Marine Technol Soc J2011; 45: 62–74.
21.
DuraEZhangYLiaoX, et al.Active learning for detection of mine-like objects in side-scan sonar imagery. IEEE J Oceanic Eng2005; 30: 360–371.
22.
HallJJAzimi-SadjadiMRKarglSG, et al.Underwater unexploded ordnance (UXO) classification using a matched subspace classifier with adaptive dictionaries. IEEE J Oceanic Eng2019; 44: 739–752.
23.
ClarkDEBellJ. Bayesian multiple target tracking in forward scan sonar images using the PHD filter. IEE Proc Radar, Sonar Navig2005; 152: 327–334.
24.
WilliamsDP. The mondrian detection algorithm for sonar imagery. IEEE Trans Geosci Remote Sens2018; 56: 1091–1102.
25.
ZhuJYuSHanZ, et al.Underwater object recognition using transformable template matching based on prior knowledge. Mathemat Probl Eng2019; 2019: 2892975.
26.
DzieciuchIGebhardtDBarngroverC, et al.Non-linear convolutional neural network for automatic detection of mine-like objects in sonar imagery. In: In V, Longhini P, Palacios A (eds) Proceedings of the 4th international conference on applications in nonlinear dynamics (ICAND 2016), 2017, pp.309–14. Cham: Springer International Publishing.
27.
AncutiCAncutiCOHaberT, et al.Enhancing underwater images and videos by fusion. In: 2012 IEEE conference on computer vision and pattern recognition, 2012, pp.81–8.
28.
Abdul GhaniASMat IsaNA. Underwater image quality enhancement through composition of dual-intensity images and rayleigh-stretching. SpringerPlus2014; 3: 1–14.
29.
LiYZhangYXuX, et al.Dust removal from high turbid underwater images using convolutional neural networks. Opt Laser Technol2019; 110: 2–6.
30.
LiCGuoCRenW, et al.An underwater image enhancement benchmark dataset and beyond. IEEE Trans Image Process2020; 29: 4376–4389.
31.
AvolaDCannistraciICascioM, et al.Real-Time GAN-Based model for underwater image enhancement. In: Foresti GL, Fusiello A and Hancock E (eds) Image analysis and processing – ICIAP 2023, 2023, pp.412–23. Cham: Springer Nature Switzerland.
32.
BoudhaneMNsiriB. Underwater image processing method for fish localization and detection in submarine environment. J Vis Commun Image Represent2016; 39: 226–238.
33.
QinHLiXLiangJ, et al.DeepFish: accurate underwater live fish recognition with a deep architecture. Neurocomputing2016; 187: 49–58.
34.
SunXShiJLiuL, et al.Transferring deep knowledge for object recognition in low-quality underwater videos. Neurocomputing2018; 275: 897–908.
35.
WangXOuyangJLiD, et al.Underwater object recognition based on deep encoding-decoding network. J Ocean Univ China2019; 18: 376–382.
36.
De NardinAZottinSPiciarelliC, et al.Few-shot pixel-precise document layout segmentation via dynamic instance generation and local thresholding. Int J Neural Syst2023; 33: 2350052.
37.
De NardinAZottinSPaierM, et al.Efficient few-shot learning for pixel-precise handwritten document layout analysis. In: Proceedings of the IEEE/CVF winter conference on applications of computer vision, 2023, pp.3680–8.
38.
ZottinSDe NardinAColombiE, et al.U-DIADS-bib: a full and few-shot pixel-precise dataset for document layout analysis of ancient manuscripts. Neural Comput Appl2024; 36: 11777–11789.
39.
JiaoPYeXZhangC, et al.Vision-based real-time marine and offshore structural health monitoring system using underwater robots. Comput Aid Civil Infrast Eng2024; 39: 281–299.
40.
MohanVSimskeSJ. Cross-sensor vision system for maritime object detection. Front Marine Sci2023; 10: 1112955.
41.
GrossetJNdaoAFougèresAJ, et al.A cooperative approach to avoiding obstacles and collisions between autonomous industrial vehicles in a simulation platform. Integr Comput Aided Eng2023; 30: 19–40.
42.
GrossetJFougèresAJDjoko-KouamM, et al.Multi-agent simulation of autonomous industrial vehicle fleets: Towards dynamic task allocation in V2X cooperation mode. Integr Comput Aided Eng2024; 31: 249–266.
43.
JeonMMoonJJeongS, et al.Autonomous flight strategy of an unmanned aerial vehicle with multimodal information for autonomous inspection of overhead transmission facilities. Comput Aid Civil Infrast Eng2024; 39: 2159–2186.
ZhangBJiDLiuS, et al.Autonomous underwater vehicle navigation: A review. Ocean Eng2023; 273: 113861.
46.
XanthidisMKalaitzakisMKarapetyanN, et al.AquaVis: a perception-aware autonomous navigation framework for underwater vehicles. In: 2021 IEEE/RSJ international conference on intelligent robots and systems (IROS); 2021. p. 5410–7.
47.
XuSZhangMSongW, et al.A systematic review and analysis of deep learning-based underwater object detection. Neurocomputing2023; 527: 204–232.
48.
González-GarcíaJGómez-EspinosaAGarcía-ValdovinosLG, et al.Experimental validation of a model-free high-order sliding mode controller with finite-time convergence for trajectory tracking of autonomous underwater vehicles. Sensors2022; 22: 488.Available from: https://www.mdpi.com/1424-8220/22/2/488.
DuttaAZissermanA. The VIA annotation software for images, audio and video. In: Proceedings of the 27th ACM international conference on multimedia. MM ’19. New York, NY, USA: ACM, 2019. Available from: https://doi.org/10.1145/3343031.3350535.
55.
BuslaevAParinovAKhvedchenyaE, et al.Albumentations: fast and flexible image augmentations. ArXiv e-prints. 2018.
56.
TavarisDScandinoLForestiGL, et al.Web Appendix: automation videos and UXO detection videos using YOLOv5s; 2025. Available at: https://lambda-iot.uniud.it/uxo_yolov5s (Online; accessed 19 February 2025).
57.
Colreavy-DonnellySCaraffiniFKuhnS, et al.Shallow buried improvised explosive device detection via convolutional neural networks. Integr Comput Aided Eng2020; 27: 403–416.
58.
ScagnettoIBrajnikGGusP, et al.Oceanus: a context-aware low-cost navigation aid for yacht racing [article]. J Navig2021; 74: 738-–749.
59.
OdettiABruzzoneGFerrettiR, et al.lake environmental data harvester (LED) for alpine Lake monitoring with autonomous surface vehicles (ASVs). Remote Sens (Basel)2024; 16: 1998.
60.
TomaACecchinatoNDrioliC, et al.Onboard audio and video processing for secure detection, localization, and tracking in counter-UAV applications. Procedia Comput Sci2022; 205: 20–27.
61.
GabrielliLMigliorelliLCantariniM, et al.An advanced multimodal driver-assistance prototype for emergency-vehicle detection. Integr Comput Aided Eng2024; 31: 381–399.
62.
RafieiMHGauthierLVAdeliH, et al.Self-supervised learning for electroencephalography. IEEE Trans Neural Netw Learn Syst2024; 35: 1457–1471.
63.
RafieiMHGauthierLVAdeliH, et al.Self-supervised learning for near-wild cognitive workload estimation. J Med Syst2024; 48: 107.
