Underwater structures are highly susceptible to damage from various factors, such as corrosion, erosion, and scouring, which can significantly reduce their load-bearing capacity. Detecting structural damage in these environments is crucial. Computer vision technology has been widely applied to the damage detection of infrastructures because of its simple operation and noncontact advantages. The captured images of underwater structures are often blurred due to the complexity of the underwater environment, significantly affecting damage detection. This study proposes a robust vision-based method for damage detection in underwater structures. Specifically, considering the potential uncertainty, the PUIE-Net model is introduced to generate multiple enhanced images and construct the corresponding probability distribution. The generated images with the highest probability distribution are selected as the optimal enhanced images, ensuring that the enhanced images are similar to the real waterless images. Subsequently, an attention mechanism is introduced to improve the YOLOv11 model. Additionally, dataset augmentation techniques are employed to simulate various underwater environments to expand the datasets for model training. Field experiments are conducted on an actual concrete bridge to validate the effectiveness of the proposed method. The damage images of the underwater structures enhanced by the proposed method are clearer than those enhanced by other methods. The proposed improved YOLOv11 model detects structural damage in enhanced underwater images with higher accuracy than the original model. In addition, the proposed method demonstrates strong robustness in detecting damage in underwater structures, even under varying camera shooting angles, small-scale damage, and camera shaking.
TengSLiuAYeX, et al. Review of intelligent detection and health assessment of underwater structures. Eng Struct2024; 308: 117958.
2.
TanHZhengLMaC, et al. Deep learning-assisted high-resolution sonar detection of local damage in underwater structures. Autom Constr2024; 164: 105479.
3.
CuiBWangCLiY, et al. Image enhancement-based detection of concrete cracks under turbid water bodies. Archit Eng Des Manag2024; 20(6): 1–22.
4.
ThiBPHNguyenVDVuongQV, et al. Enhancing the load-bearing capacity of a damaged hydraulic structure rehabilitated by underwater concreting. Eng Technol Appl Sci Res2024; 14(4): 15651–15655.
5.
TengSLiuAWuZ, et al. Automated detection of underwater cracks based on fusion of optical and texture information. Eng Struct2024; 315: 118515.
6.
WangXXuHWangJ, et al. Damage area identification of underwater structures based on wavelength difference of light source. J Civ Struct Health Monit2022; 12(3): 501–515.
7.
PushpakumaraBHJThusithaGA. Development of a structural health monitoring tool for underwater concrete structures. J Constr Eng Manag2021; 147(10): 04021135.
8.
WuZLiuATengS, et al. A multi-stage method for defect detection of underwater structures based on deep learning. Struct Health Monit. Epub ahead of print 25 December 2024. DOI: 10.1177/14759217241301098.
9.
LiWChenGGeJ, et al. High sensitivity rotating alternating current field measurement for arbitrary-angle underwater cracks. NDT & E Int2016; 79: 123–131.
10.
QiZLLiuDZhangJ, et al. Micro-concrete crack detection of underwater structures based on convolutional neural network. Mach Vis Appl2022; 33(5): 74.
11.
HouSShenHWuT, et al. Underwater surface defect recognition of bridges based on fusion of semantic segmentation and three-dimensional point cloud. J Bridge Eng2025; 30(1): 04024101.
12.
SunWHouSFanJ, et al. Ultrasonic computed tomography-based internal-defect detection and location of underwater concrete piers. Smart Mater Struct2023; 32(12): 125021.
13.
ChenDHuangBKangF.A review of detection technologies for underwater cracks on concrete dam surfaces. Appl Sci2023; 13(6): 3564.
14.
WangHLiWYinX, et al. A flexible alternating current field measurement magnetic sensor array for in-situ inspection of cracks in underwater structure. IEEE Trans Instrum Meas2024; 73: 1–10.
15.
ZhuYChenJZhangY, et al. Temperature tracer method for crack detection in underwater concrete structures. Struct Control Health Monit2020; 27(9): e2595.
16.
LiYZhaoHWeiY, et al. Vision-guided crack identification and size quantification framework for dam underwater concrete structures. Struct Health Monit2025; 24(4): 2125–2148.
17.
WanHPZhangWJChenY, et al. An efficient three-dimensional point cloud segmentation method for the dimensional quality assessment of precast concrete components utilizing multi-view information fusion. J Comput Civ Eng2025; 39(3): 04025028.
18.
HouSJiaoDDongB, et al. Underwater inspection of bridge substructures using sonar and deep convolutional network. Adv Eng Inform2022; 52: 101545.
19.
GeLSinghPSadhuA.Advanced deep learning framework for underwater object detection with multibeam forward-looking sonar. Struct Health Monit2025 ; 24(4): 1991–2007
20.
SunWHouSWuG, et al. Turbid image tackling framework towards underwater concrete bridge detection based on distance control and deep learning. Adv Eng Inform2024; 62: 102723.
21.
HoangNDNguyenQL.Computer vision-based recognition of pavement crack patterns using light gradient boosting machine, deep neural network, and convolutional neural network. J Soft Comput Civ Eng2023; 7(3): 21–51.
22.
LiaoWZhangSWuY, et al. 9999Research on intelligent damage detection of far-sea cage based on machine vision and deep learning. Aquacult Eng2022; 96: 102219.
23.
ChenCPWangJZouL, et al. Underwater dam image crack segmentation based on mathematical morpholog. Appl Mech Mater2012; 220: 1315–1319.
24.
ShiPFanXNiJ, et al. A detection and classification approach for underwater dam cracks. Struct Health Monit2016; 15(5): 541–554.
25.
WangJLiuH.Defect detection method of underwater bored cast-in-place pile based on optical image in borehole. J Civ Struct Health Monit2024; 14(1): 189–207.
26.
ChenXWuGHouS, et al. Development of tactile imaging for damage detection of the underwater structure. Sensors2019; 19(18): 3925.
27.
ShiPShaoSLiuY, et al. CrackInst: a real-time instance segmentation method for underwater dam cracks. IEEE Trans Instrum Meas2024; 73: 1–11.
28.
HuangYZhuoQFuJ, et al. Research on evaluation method of underwater image quality and performance of underwater structure defect detection model. Eng Struct2024; 306: 117797.
29.
ZhangHLiJKangF, et al. Monitoring depth and width of cracks in underwater concrete structures using embedded smart aggregates. Measurement2022; 204: 112078.
30.
BolourianNNasrollahiMBahreiniF, et al. Point cloud-based concrete surface defect semantic segmentation. J Comput Civ Eng2023; 37(2): 04022056.
31.
WanHPZhangWJGeHB, et al. Improved vision-based method for detection of unauthorized intrusion by construction sites workers. J Constr Eng Manag2023; 149(7): 04023040.
32.
LiYBaoTLiT, et al. A robust real-time method for identifying hydraulic tunnel structural defects using deep learning and computer vision. Comput Aided Civ Infrastruct Eng2023; 38(10): 1381–1399.
33.
CaoWLiJ.A novel image multitasking enhancement model for underwater crack detection. Struct Health Monit2025; 24(4): 1969–1990
34.
YeXLuoKWangH, et al. An advanced AI-based lightweight two-stage damage detection of the underwater structure model. Adv Eng Inform2024; 62: 102553.
35.
HuKChenZKangH, et al. 3D vision technologies for a self-developed structural external crack damage recognition robot. Autom Constr2024; 159: 105262.
36.
WanHPZhuYKLuoY, et al. Unsupervised deep learning approach for structural anomaly detection using probabilistic features. Struct Health Monit2025; 24(1): 3–33.
37.
PhamNLTaQBKimJT.Pseudo-damage simulation and CNN deep learning for d999999amage identification of submerged structure-foundation systems. Struct Health Monit2025. DOI: 10.1177/14759217241288773.
38.
JiangSWangWSuZ, et al. Automatic detection of surface defects on underwater pile-pier of bridges based on image fusion and deep learning. Struct Control Health Monit2023; 2023(1): 8429099.
39.
CuiBWangCLiY, et al. Application of computer vision techniques to damage detection in underwater concrete structures. Alex Eng J2024; 104: 745–752.
40.
QianJLiJWangY, et al. Underwater image recovery method based on hyperspectral polarization imaging. Opt Commun2021; 484: 126691.
41.
XinGFanXShiP, et al. A fine extraction algorithm for image-based surface cracks in underwater dams. Meas Sci Technol2022; 34(3): 035402.
42.
ZhengZHuangXWangL.Underwater low-light enhancement network based on bright channel prior and attention mechanism. PLoS One2023; 18(2): e0281093.
43.
HuangBKangFLiX, et al. Underwater dam crack image generation based on unsupervised image-to-image translation. Autom Constr2024; 163: 105430.
44.
LiXSunHSongT, et al. A method of underwater bridge structure damage detection method based on a lightweight deep convolutional network. IET Image Process2022; 16(14): 3893–3909.
45.
WuZHuZToddMD, et al. Automated crack detection for underwater inspection of miter gates with unmanned underwater vehicle. Struct Health Monit2023; 2023: 706–712.
46.
WuZHuZToddMD. Uncertainty quantification for deep learning–based automatic crack detection in the underwater environment. In: IMAC, a conference and exposition on structural dynamics, November 2024, pp. 133–136. Cham: Springer Nature Switzerland. https://doi.org/10.1007/978-3-031-68893-5_20
47.
ChenDKangFLiJ, et al. Enhancement of underwater dam crack images using multi-feature fusion. Autom Constr2024; 167: 105727.
48.
FuZWangWHuangY, et al. Uncertainty inspired underwater image enhancement. In: European conference on computer vision, October 2022, pp. 465–482. Cham: Springer Nature Switzerland. https://doi.org/10.1007/978-3-031-19797-0_27
49.
KhanamRHussainM.Yolov11: an overview of the key architectural enhancements. arXiv preprint. arXiv: 2410.17725, 2024.
50.
LiuYShaoZHoffmannN.Global attention mechanism: retain information to enhance channel-spatial interactions. arXiv preprint. arXiv:2112.05561, 2021.
51.
YangMSowmyaA.An underwater color image quality evaluation metric. IEEE Trans Image Process2015; 24(12): 6062–6071.
TengSLiuASituZ, et al. Plug-and-play method for segmenting concrete bridge cracks using the segment anything model with a fractal dimension matrix prompt. Autom Constr2025; 170: 105906.
