{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,1,14]],"date-time":"2026-01-14T16:06:39Z","timestamp":1768406799959,"version":"3.49.0"},"reference-count":42,"publisher":"MDPI AG","issue":"17","license":[{"start":{"date-parts":[[2024,8,28]],"date-time":"2024-08-28T00:00:00Z","timestamp":1724803200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"technology project funding from Shandong Zhiyuan Electric Power Design Consulting Co., Ltd.","award":["ZY-2023-04"],"award-info":[{"award-number":["ZY-2023-04"]}]},{"name":"technology project funding from Shandong Zhiyuan Electric Power Design Consulting Co., Ltd.","award":["52208223"],"award-info":[{"award-number":["52208223"]}]},{"name":"technology project funding from Shandong Zhiyuan Electric Power Design Consulting Co., Ltd.","award":["2023JJ40754"],"award-info":[{"award-number":["2023JJ40754"]}]},{"name":"technology project funding from Shandong Zhiyuan Electric Power Design Consulting Co., Ltd.","award":["2022M713545"],"award-info":[{"award-number":["2022M713545"]}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["ZY-2023-04"],"award-info":[{"award-number":["ZY-2023-04"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["52208223"],"award-info":[{"award-number":["52208223"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["2023JJ40754"],"award-info":[{"award-number":["2023JJ40754"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["2022M713545"],"award-info":[{"award-number":["2022M713545"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Hunan Provincial Natural Science Foundation","award":["ZY-2023-04"],"award-info":[{"award-number":["ZY-2023-04"]}]},{"name":"Hunan Provincial Natural Science Foundation","award":["52208223"],"award-info":[{"award-number":["52208223"]}]},{"name":"Hunan Provincial Natural Science Foundation","award":["2023JJ40754"],"award-info":[{"award-number":["2023JJ40754"]}]},{"name":"Hunan Provincial Natural Science Foundation","award":["2022M713545"],"award-info":[{"award-number":["2022M713545"]}]},{"name":"China Postdoctoral Science Foundation","award":["ZY-2023-04"],"award-info":[{"award-number":["ZY-2023-04"]}]},{"name":"China Postdoctoral Science Foundation","award":["52208223"],"award-info":[{"award-number":["52208223"]}]},{"name":"China Postdoctoral Science Foundation","award":["2023JJ40754"],"award-info":[{"award-number":["2023JJ40754"]}]},{"name":"China Postdoctoral Science Foundation","award":["2022M713545"],"award-info":[{"award-number":["2022M713545"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Sensors"],"abstract":"<jats:p>Water leakage defects often occur in underground structures, leading to accelerated structural aging and threatening structural safety. Leakage identification can detect early diseases of underground structures and provide important guidance for reinforcement and maintenance. Deep learning-based computer vision methods have been rapidly developed and widely used in many fields. However, establishing a deep learning model for underground structure leakage identification usually requires a lot of training data on leakage defects, which is very expensive. To overcome the data shortage, a deep neural network method for leakage identification is developed based on transfer learning in this paper. For comparison, four famous classification models, including VGG16, AlexNet, SqueezeNet, and ResNet18, are constructed. To train the classification models, a transfer learning strategy is developed, and a dataset of underground structure leakage is created. Finally, the classification performance on the leakage dataset of different deep learning models is comparatively studied under different sizes of training data. The results showed that the VGG16, AlexNet, and SqueezeNet models with transfer learning can overall provide higher and more stable classification performance on the leakage dataset than those without transfer learning. The ResNet18 model with transfer learning can overall provide a similar value of classification performance on the leakage dataset than that without transfer learning, but its classification performance is more stable than that without transfer learning. In addition, the SqueezeNet model obtains an overall higher and more stable performance than the comparative models on the leakage dataset for all classification metrics.<\/jats:p>","DOI":"10.3390\/s24175569","type":"journal-article","created":{"date-parts":[[2024,8,28]],"date-time":"2024-08-28T07:52:08Z","timestamp":1724831528000},"page":"5569","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":3,"title":["Leakage Identification of Underground Structures Using Classification Deep Neural Networks and Transfer Learning"],"prefix":"10.3390","volume":"24","author":[{"given":"Wenyang","family":"Wang","sequence":"first","affiliation":[{"name":"Shandong Zhiyuan Electric Power Design Consulting Co., Ltd., Jinan 250021, China"},{"name":"Economic & Technology Research Institute of State Grid Shandong Electric Power Company, Jinan 250021, China"}]},{"given":"Qingwei","family":"Chen","sequence":"additional","affiliation":[{"name":"Shandong Zhiyuan Electric Power Design Consulting Co., Ltd., Jinan 250021, China"},{"name":"Economic & Technology Research Institute of State Grid Shandong Electric Power Company, Jinan 250021, China"}]},{"given":"Yongjiang","family":"Shen","sequence":"additional","affiliation":[{"name":"Hunan Province Key Laboratory for Disaster Prevention and Mitigation of Rail Transit Engineering Structure, Central South University, Changsha 410075, China"},{"name":"School of Civil Engineering, Central South University, Changsha 410075, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9676-3229","authenticated-orcid":false,"given":"Zhengliang","family":"Xiang","sequence":"additional","affiliation":[{"name":"Hunan Province Key Laboratory for Disaster Prevention and Mitigation of Rail Transit Engineering Structure, Central South University, Changsha 410075, China"},{"name":"School of Civil Engineering, Central South University, Changsha 410075, China"}]}],"member":"1968","published-online":{"date-parts":[[2024,8,28]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"109423","DOI":"10.1016\/j.ress.2023.109423","article-title":"Development of a non-Gaussian copula Bayesian network for safety assessment of metro tunnel maintenance","volume":"238","author":"Liu","year":"2023","journal-title":"Reliab. Eng. Syst. Saf."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"109784","DOI":"10.1016\/j.buildenv.2022.109784","article-title":"Natural ventilation design for underground parking garages","volume":"227","author":"Chu","year":"2023","journal-title":"Build. Environ."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"107627","DOI":"10.1016\/j.epsr.2021.107627","article-title":"Numerical analysis and safety design of grounding systems in underground compact substations","volume":"203","author":"Colominas","year":"2022","journal-title":"Electr. Power Syst. Res."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"1353","DOI":"10.1177\/1475921720972416","article-title":"Machine learning paradigm for structural health monitoring","volume":"20","author":"Bao","year":"2021","journal-title":"Struct. Health Monit."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"234","DOI":"10.1016\/j.eng.2018.11.027","article-title":"The state of the art of data science and engineering in structural health monitoring","volume":"5","author":"Bao","year":"2019","journal-title":"Engineering"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"1339","DOI":"10.1007\/s13349-024-00788-8","article-title":"A tunnel structure health monitoring method based on surface strain monitoring","volume":"14","author":"Zhou","year":"2024","journal-title":"J. Civ. Struct. Health Monit."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"103954","DOI":"10.1016\/j.jobe.2021.103954","article-title":"Structural health monitoring of civil engineering structures by using the internet of things: A review","volume":"48","author":"Mishra","year":"2022","journal-title":"J. Build. Eng."},{"key":"ref_8","doi-asserted-by":"crossref","first-page":"104464","DOI":"10.1016\/j.autcon.2022.104464","article-title":"Construction and maintenance of urban underground infrastructure with digital technologies","volume":"141","author":"Wang","year":"2022","journal-title":"Autom. Constr."},{"key":"ref_9","doi-asserted-by":"crossref","first-page":"886","DOI":"10.1016\/j.jrmge.2022.06.015","article-title":"Real-time prediction of mechanical behaviors of underwater shield tunnel structure using machine learning method based on structural health monitoring data","volume":"15","author":"Tan","year":"2023","journal-title":"J. Rock Mech. Geotech. Eng."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"1184","DOI":"10.1177\/14759217231183661","article-title":"An active learning method for crack detection based on subset searching and weighted sampling","volume":"23","author":"Xiang","year":"2024","journal-title":"Struct. Health Monit."},{"key":"ref_11","doi-asserted-by":"crossref","first-page":"692","DOI":"10.1177\/1475921720935585","article-title":"A review of computer vision-based structural health monitoring at local and global levels","volume":"20","author":"Dong","year":"2021","journal-title":"Struct. Health Monit."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"401","DOI":"10.1177\/1475921718757405","article-title":"Computer vision and deep learning-based data anomaly detection method for structural health monitoring","volume":"18","author":"Bao","year":"2019","journal-title":"Struct. Health Monit."},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"14","DOI":"10.1111\/mice.12550","article-title":"Automated crack evaluation of a high-rise bridge pier using a ring-type climbing robot","volume":"36","author":"Jang","year":"2021","journal-title":"Comput.-Aided Civ. Infrastruct. Eng."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"e2381","DOI":"10.1002\/stc.2381","article-title":"Image-based concrete crack assessment using mask and region-based convolutional neural network","volume":"26","author":"Kim","year":"2019","journal-title":"Struct. Control Health Monit."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"104435","DOI":"10.1016\/j.autcon.2022.104435","article-title":"Excavator joint node-based pose estimation using lightweight fully convolutional network","volume":"141","author":"Guo","year":"2022","journal-title":"Autom. Constr."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"105478","DOI":"10.1016\/j.engappai.2022.105478","article-title":"Computer vision framework for crack detection of civil infrastructure\u2014A review","volume":"117","author":"Ai","year":"2023","journal-title":"Eng. Appl. Artif. Intell."},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"3105","DOI":"10.1109\/TETCI.2024.3369999","article-title":"WLR-Net: An Improved YOLO-V7 With Edge Constraints and Attention Mechanism for Water Leakage Recognition in the Tunnel","volume":"8","author":"Chen","year":"2024","journal-title":"IEEE Trans. Emerg. Top. Comput. Intell."},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Li, J., He, Z., Li, D., and Zheng, A. (2022). Research on water seepage detection technology of tunnel asphalt pavement based on deep learning and digital image processing. Sci. Rep., 12.","DOI":"10.1038\/s41598-022-15828-w"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"32471","DOI":"10.1109\/ACCESS.2018.2842797","article-title":"Detection of water leakage in underground tunnels using corrected intensity data and 3D point cloud of terrestrial laser scanning","volume":"6","author":"Xu","year":"2018","journal-title":"IEEE Access"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"166","DOI":"10.1016\/j.tust.2018.04.002","article-title":"Deep learning based image recognition for crack and leakage defects of metro shield tunnel","volume":"77","author":"Huang","year":"2018","journal-title":"Tunn. Undergr. Space Technol."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"823","DOI":"10.1080\/01431160600746456","article-title":"A survey of image classification methods and techniques for improving classification performance","volume":"28","author":"Lu","year":"2007","journal-title":"Int. J. Remote Sens."},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"257","DOI":"10.1109\/JPROC.2023.3238524","article-title":"Object detection in 20 years: A survey","volume":"111","author":"Zou","year":"2023","journal-title":"Proc. IEEE"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"302","DOI":"10.1016\/j.neucom.2019.11.118","article-title":"A brief survey on semantic segmentation with deep learning","volume":"406","author":"Hao","year":"2020","journal-title":"Neurocomputing"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"171","DOI":"10.1007\/s13735-020-00195-x","article-title":"A survey on instance segmentation: State of the art","volume":"9","author":"Hafiz","year":"2020","journal-title":"Int. J. Multimed. Inf. Retr."},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Qassim, H., Verma, A., and Feinzimer, D. (2018, January 8\u201310). Compressed residual-VGG16 CNN model for big data places image recognition. Proceedings of the 2018 IEEE 8th Annual Computing and Communication Workshop and Conference (CCWC), Las Vegas, NV, USA.","DOI":"10.1109\/CCWC.2018.8301729"},{"key":"ref_26","unstructured":"Iandola, F.N., Han, S., Moskewicz, M.W., Ashraf, K., Dally, W.J., and Keutzer, K. (2016). SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5 MB model size. arXiv."},{"key":"ref_27","doi-asserted-by":"crossref","unstructured":"Lee, H.J., Ullah, I., Wan, W., Gao, Y., and Fang, Z. (2019). Real-time vehicle make and model recognition with the residual SqueezeNet architecture. Sensors, 19.","DOI":"10.3390\/s19050982"},{"key":"ref_28","unstructured":"Targ, S., Almeida, D., and Lyman, K. (2016). Resnet in resnet: Generalizing residual architectures. arXiv."},{"key":"ref_29","doi-asserted-by":"crossref","first-page":"1037","DOI":"10.1080\/15732479.2019.1680709","article-title":"Vision-based defects detection for bridges using transfer learning and convolutional neural networks","volume":"16","author":"Zhu","year":"2020","journal-title":"Struct. Infrastruct. Eng."},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"3077","DOI":"10.1177\/14759217221142174","article-title":"Transfer learning-based data anomaly detection for structural health monitoring","volume":"22","author":"Pan","year":"2023","journal-title":"Struct. Health Monit."},{"key":"ref_31","doi-asserted-by":"crossref","first-page":"14179","DOI":"10.1007\/s00521-021-06066-8","article-title":"Image classification method on class imbalance datasets using multi-scale CNN and two-stage transfer learning","volume":"33","author":"Liu","year":"2021","journal-title":"Neural Comput. Appl."},{"key":"ref_32","doi-asserted-by":"crossref","first-page":"104381","DOI":"10.1016\/j.autcon.2022.104381","article-title":"Few-shot classification of fa\u00e7ade defects based on extensible classifier and contrastive learning","volume":"141","author":"Cui","year":"2022","journal-title":"Autom. Constr."},{"key":"ref_33","doi-asserted-by":"crossref","first-page":"1494","DOI":"10.1177\/1475921720921135","article-title":"Attribute-based structural damage identification by few-shot meta learning with inter-class knowledge transfer","volume":"20","author":"Xu","year":"2021","journal-title":"Struct. Health Monit."},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"302","DOI":"10.1111\/mice.12632","article-title":"Semi-supervised learning based on convolutional neural network and uncertainty filter for fa\u00e7ade defects classification","volume":"36","author":"Guo","year":"2021","journal-title":"Comput.-Aided Civ. Infrastruct. Eng."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"104139","DOI":"10.1016\/j.autcon.2022.104139","article-title":"Road damage detection using super-resolution and semi-supervised learning with generative adversarial network","volume":"135","author":"Shim","year":"2022","journal-title":"Autom. Constr."},{"key":"ref_36","doi-asserted-by":"crossref","first-page":"2408","DOI":"10.1111\/mice.12998","article-title":"Informativeness-guided active learning for deep learning-based fa\u00e7ade defects detection","volume":"38","author":"Guo","year":"2023","journal-title":"Comput.-Aided Civ. Infrastruct. Eng."},{"key":"ref_37","doi-asserted-by":"crossref","first-page":"1089","DOI":"10.1111\/mice.12851","article-title":"Multistage semisupervised active learning framework for crack identification, segmentation, and measurement of bridges","volume":"37","author":"Zheng","year":"2022","journal-title":"Comput.-Aided Civ. Infrastruct. Eng."},{"key":"ref_38","first-page":"195","article-title":"An active learning method with difficulty learning mechanism for crack detection","volume":"29","author":"Shu","year":"2022","journal-title":"Smart Struct. Syst."},{"key":"ref_39","doi-asserted-by":"crossref","unstructured":"Man, K., Liu, R., Liu, X., Song, Z., Liu, Z., Cao, Z., and Wu, L. (2022). Water leakage and crack identification in tunnels based on transfer-learning and convolutional neural networks. Water, 14.","DOI":"10.3390\/w14091462"},{"key":"ref_40","doi-asserted-by":"crossref","unstructured":"Zoubir, H., Rguig, M., El Aroussi, M., Chehri, A., Saadane, R., and Jeon, G. (2022). Concrete bridge defects identification and localization based on classification deep convolutional neural networks and transfer learning. Remote Sens., 14.","DOI":"10.3390\/rs14194882"},{"key":"ref_41","doi-asserted-by":"crossref","first-page":"103382","DOI":"10.1016\/j.autcon.2020.103382","article-title":"Inspection of surface defects on stay cables using a robot and transfer learning","volume":"119","author":"Hou","year":"2020","journal-title":"Autom. Constr."},{"key":"ref_42","doi-asserted-by":"crossref","first-page":"103524","DOI":"10.1016\/j.tust.2020.103524","article-title":"Deep learning-based automatic recognition of water leakage area in shield tunnel lining","volume":"104","author":"Xue","year":"2020","journal-title":"Tunn. Undergr. Space Technol."}],"container-title":["Sensors"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1424-8220\/24\/17\/5569\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,10]],"date-time":"2025-10-10T15:44:15Z","timestamp":1760111055000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1424-8220\/24\/17\/5569"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,8,28]]},"references-count":42,"journal-issue":{"issue":"17","published-online":{"date-parts":[[2024,9]]}},"alternative-id":["s24175569"],"URL":"https:\/\/doi.org\/10.3390\/s24175569","relation":{},"ISSN":["1424-8220"],"issn-type":[{"value":"1424-8220","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,8,28]]}}}