{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,12,22]],"date-time":"2025-12-22T12:23:40Z","timestamp":1766406220465,"version":"3.48.0"},"reference-count":44,"publisher":"MDPI AG","issue":"1","license":[{"start":{"date-parts":[[2025,12,22]],"date-time":"2025-12-22T00:00:00Z","timestamp":1766361600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"name":"Hubei Key Research and Development Program","award":["2023BCB045"],"award-info":[{"award-number":["2023BCB045"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["BDCC"],"abstract":"<jats:p>Bridge health diagnosis plays a vital role in ensuring structural safety and extending service life while reducing maintenance costs. Traditional structural health monitoring approaches rely on sensor-based measurements, which are costly, labor-intensive, and limited in coverage. To address these challenges, we propose a three-phase solution that integrates the Dynamic Lightweight Vision-Language Model (DL-VLM), domain adaptation, and knowledge-enhanced reasoning. First, as the core of the framework, the DL-VLM consists of three components: a visual information encoder with multi-scale feature selection, a text encoder for processing inspection-related language, and a multimodal alignment module. Second, to enhance practical applicability, we further introduce domain-specific fine-tuning on the Bridge-SHM dataset, enabling the model to acquire specialized knowledge of bridge construction, defects, and structural components. Third, a knowledge retrieval augmentation module is incorporated, leveraging external knowledge graphs and vector-based retrieval to provide contextually relevant information and improve diagnostic reasoning. Experiments on high-resolution bridge inspection datasets demonstrate that DL-VLM achieves competitive diagnostic accuracy while substantially reducing computational cost. The combination of domain-specific fine-tuning and knowledge augmentation significantly improves performance on specialized tasks, supporting efficient and practical deployment in real-world structural health monitoring scenarios.<\/jats:p>","DOI":"10.3390\/bdcc10010003","type":"journal-article","created":{"date-parts":[[2025,12,22]],"date-time":"2025-12-22T11:43:31Z","timestamp":1766403811000},"page":"3","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["DL-VLM: A Dynamic Lightweight Vision-Language Model for Bridge Health Diagnosis"],"prefix":"10.3390","volume":"10","author":[{"given":"Shenghao","family":"Liang","sequence":"first","affiliation":[{"name":"School of Computer Science, Wuhan University, Wuhan 430072, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhiheng","family":"He","sequence":"additional","affiliation":[{"name":"School of Computer Science, Wuhan University, Wuhan 430072, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hao","family":"Gui","sequence":"additional","affiliation":[{"name":"School of Computer Science, Wuhan University, Wuhan 430072, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Feng","family":"Liu","sequence":"additional","affiliation":[{"name":"School of Computer Science, Wuhan University, Wuhan 430072, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2025,12,22]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"111177","DOI":"10.1016\/j.ymssp.2024.111177","article-title":"Bridge damage localization and quantification using deep learning and FEM static simulation","volume":"211","author":"Sun","year":"2024","journal-title":"Mech. Syst. Signal Process."},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Ngeljaratan, L., Moustafa, M.A., Sumarno, A., Prasetyo, A.M., Sari, D.P., and Maidina, M. (2024). Improved blob-based feature detection and refined matching algorithms for seismic structural health monitoring of bridges using a vision-based sensor system. Infrastructures, 9.","DOI":"10.3390\/infrastructures9060097"},{"key":"ref_3","doi-asserted-by":"crossref","unstructured":"D\u2019Angelo, M., Civera, M., Giordano, P.F., Borlenghi, P., Ballio, F., Limongelli, M.P., and Chiaia, B. (2025). Bridge collapses in Italy across the 21st century: Survey and statistical analysis. Struct. Infrastruct. Eng., 1\u201323.","DOI":"10.1080\/15732479.2025.2483500"},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"17","DOI":"10.1007\/s43503-025-00063-9","article-title":"Semantic and lexical analysis of pre-trained vision language artificial intelligence models for automated image descriptions in civil engineering","volume":"4","author":"Bazrafshan","year":"2025","journal-title":"AI Civ. Eng."},{"key":"ref_5","doi-asserted-by":"crossref","unstructured":"Micozzi, F., Morici, M., Zona, A., and Dall\u2019Asta, A. (2023). Vision-based structural monitoring: Application to a medium-span post-tensioned concrete bridge under vehicular traffic. Infrastructures, 8.","DOI":"10.3390\/infrastructures8100152"},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"70","DOI":"10.1007\/s10921-020-00715-z","article-title":"Scale\u2013space data augmentation for deep transfer learning of crack damage from small sized datasets","volume":"39","author":"Tang","year":"2020","journal-title":"J. Nondestruct. Eval."},{"key":"ref_7","doi-asserted-by":"crossref","first-page":"383","DOI":"10.1016\/j.neucom.2019.08.107","article-title":"Automatic crack distress classification from concrete surface images using a novel deep-width network architecture","volume":"397","author":"Guo","year":"2020","journal-title":"Neurocomputing"},{"key":"ref_8","unstructured":"Zhu, D., Chen, J., Shen, X., Li, X., and Elhoseiny, M. (2023). Minigpt-4: Enhancing vision-language understanding with advanced large language models. arXiv."},{"key":"ref_9","unstructured":"Li, J., Li, D., Savarese, S., and Hoi, S. (2023, January 23\u201329). Blip-2: Bootstrapping language-image pre-training with frozen image encoders and large language models. Proceedings of the International Conference on Machine Learning, Honolulu, HI, USA."},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"012103","DOI":"10.1088\/1742-6596\/1732\/1\/012103","article-title":"Bridge health monitoring and evaluation system","volume":"Volume 768","author":"Yang","year":"2021","journal-title":"Proceedings of the IOP Conference Series: Earth and Environmental Science"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Fu, Y., Zhu, Y., Hoang, T., Mechitov, K., and Spencer, B.F. (2022). xImpact: Intelligent Wireless System for Cost-Effective Rapid Condition Assessment of Bridges Under Impacts. Sensors, 22.","DOI":"10.3390\/s22155701"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Preethichandra, D., Suntharavadivel, T., Kalutara, P., Piyathilaka, L., and Izhar, U. (2023). Influence of smart sensors on structural health monitoring systems and future asset management practices. Sensors, 23.","DOI":"10.3390\/s23198279"},{"key":"ref_13","doi-asserted-by":"crossref","first-page":"2621","DOI":"10.1007\/s11831-020-09471-9","article-title":"Machine Learning Algorithms in Civil Structural Health Monitoring: A Systematic Review","volume":"28","author":"Flah","year":"2021","journal-title":"Arch. Comput. Methods Eng."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"775","DOI":"10.1111\/mice.12532","article-title":"Applicability of machine learning to a crack model in concrete bridges","volume":"35","author":"Okazaki","year":"2020","journal-title":"Comput.-Aided Civ. Infrastruct. Eng."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"674","DOI":"10.1007\/s42452-025-07272-y","article-title":"Deep learning-based crack detection and prediction for structural health monitoring","volume":"7","author":"Roy","year":"2025","journal-title":"Discov. Appl. Sci."},{"key":"ref_16","doi-asserted-by":"crossref","first-page":"105719","DOI":"10.1016\/j.autcon.2024.105719","article-title":"Artificial intelligence in structural health management of existing bridges","volume":"167","author":"Cardellicchio","year":"2024","journal-title":"Autom. Constr."},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Jim\u00e9nez Rios, A., Plevris, V., and Nogal, M. (2023). Bridge management through digital twin-based anomaly detection systems: A systematic review. Front. Built Environ., 9.","DOI":"10.3389\/fbuil.2023.1176621"},{"key":"ref_18","doi-asserted-by":"crossref","first-page":"100203","DOI":"10.1016\/j.kscej.2025.100203","article-title":"Advances in artificial intelligence for structural health monitoring: A comprehensive review","volume":"29","author":"Spencer","year":"2025","journal-title":"KSCE J. Civ. Eng."},{"key":"ref_19","unstructured":"Sharir, G., Noy, A., and Zelnik-Manor, L. (2021). An image is worth 16x16 words, what is a video worth?. arXiv."},{"key":"ref_20","unstructured":"Radford, A., Kim, J.W., Hallacy, C., Ramesh, A., Goh, G., Agarwal, S., Sastry, G., Askell, A., Mishkin, P., and Clark, J. (2021, January 18\u201324). Learning transferable visual models from natural language supervision. Proceedings of the International Conference on Machine Learning, Virtual."},{"key":"ref_21","first-page":"34892","article-title":"Visual instruction tuning","volume":"36","author":"Liu","year":"2023","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_22","doi-asserted-by":"crossref","unstructured":"Liu, H., Li, C., Li, Y., and Lee, Y.J. (2024, January 16\u201322). Improved baselines with visual instruction tuning. Proceedings of the the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.","DOI":"10.1109\/CVPR52733.2024.02484"},{"key":"ref_23","unstructured":"Li, Y. (2025). MOR-VIT: Efficient Vision Transformer with Mixture-of-Recursions. arXiv."},{"key":"ref_24","doi-asserted-by":"crossref","unstructured":"Wu, Q., Xu, W., Liu, W., Tan, T., Liujianfeng, L., Li, A., Luan, J., Wang, B., and Shang, S. (2024, January 12\u201316). Mobilevlm: A vision-language model for better intra-and inter-ui understanding. Proceedings of the Findings of the Association for Computational Linguistics: EMNLP 2024, Miami, FL, USA.","DOI":"10.18653\/v1\/2024.findings-emnlp.599"},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"Fan, H., Xiong, B., Mangalam, K., Li, Y., Yan, Z., Malik, J., and Feichtenhofer, C. (2021, January 11\u201317). Multiscale vision transformers. Proceedings of the IEEE\/CVF International Conference on Computer Vision, Virtual.","DOI":"10.1109\/ICCV48922.2021.00675"},{"key":"ref_26","unstructured":"Zhang, Y., Fan, C.K., Ma, J., Zheng, W., Huang, T., Cheng, K., Gudovskiy, D., Okuno, T., Nakata, Y., and Keutzer, K. (2024). Sparsevlm: Visual token sparsification for efficient vision-language model inference. arXiv."},{"key":"ref_27","first-page":"3","article-title":"Lora: Low-rank adaptation of large language models","volume":"1","author":"Hu","year":"2022","journal-title":"ICLR"},{"key":"ref_28","first-page":"10088","article-title":"Qlora: Efficient finetuning of quantized llms","volume":"36","author":"Dettmers","year":"2023","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_29","unstructured":"Zhang, Q., Chen, M., Bukharin, A., Karampatziakis, N., He, P., Cheng, Y., Chen, W., and Zhao, T. (2023). Adalora: Adaptive budget allocation for parameter-efficient fine-tuning. arXiv."},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Karpukhin, V., Oguz, B., Min, S., Lewis, P.S., Wu, L., Edunov, S., Chen, D., and Yih, W.t. (2020, January 16\u201320). Dense Passage Retrieval for Open-Domain Question Answering. Proceedings of the Conference on Empirical Methods in Natural Language Processing, Online.","DOI":"10.18653\/v1\/2020.emnlp-main.550"},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Reimers, N., and Gurevych, I. (2019). Sentence-bert: Sentence embeddings using siamese bert-networks. arXiv.","DOI":"10.18653\/v1\/D19-1410"},{"key":"ref_32","unstructured":"Zhu, Y., Xie, C., Liang, S., Zheng, B., and Guo, S. (2024). Focusllava: A coarse-to-fine approach for efficient and effective visual token compression. arXiv."},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"Rumelhart, D.E., Hinton, G.E., and Williams, R.J. (1985). Learning Internal Representations by Error Propagation, Institute for Cognitive Science, University of California. Technical Report ICS-8506.","DOI":"10.21236\/ADA164453"},{"key":"ref_34","doi-asserted-by":"crossref","unstructured":"Misra, D. (2019). Mish: A self regularized non-monotonic activation function. arXiv.","DOI":"10.5244\/C.34.191"},{"key":"ref_35","unstructured":"Xu, B., Wang, N., Chen, T., and Li, M. (2015). Empirical evaluation of rectified activations in convolutional network. arXiv."},{"key":"ref_36","unstructured":"Hendrycks, D. (2016). Gaussian Error Linear Units (Gelus). arXiv."},{"key":"ref_37","doi-asserted-by":"crossref","unstructured":"Hudson, D.A., and Manning, C.D. (2019, January 16\u201320). Gqa: A new dataset for real-world visual reasoning and compositional question answering. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, CA, USA.","DOI":"10.1109\/CVPR.2019.00686"},{"key":"ref_38","first-page":"2507","article-title":"Learn to explain: Multimodal reasoning via thought chains for science question answering","volume":"35","author":"Lu","year":"2022","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_39","doi-asserted-by":"crossref","unstructured":"Singh, A., Natarajan, V., Shah, M., Jiang, Y., Chen, X., Batra, D., Parikh, D., and Rohrbach, M. (2019, January 16\u201320). Towards vqa models that can read. Proceedings of the the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, CA, USA.","DOI":"10.1109\/CVPR.2019.00851"},{"key":"ref_40","doi-asserted-by":"crossref","unstructured":"Li, Y., Du, Y., Zhou, K., Wang, J., Zhao, W.X., and Wen, J.R. (2023). Evaluating object hallucination in large vision-language models. arXiv.","DOI":"10.18653\/v1\/2023.emnlp-main.20"},{"key":"ref_41","unstructured":"Liu, Y., Duan, H., Zhang, Y., Li, B., Zhang, S., Zhao, W., Yuan, Y., Wang, J., He, C., and Liu, Z. (October, January 29). Mmbench: Is your multi-modal model an all-around player?. Proceedings of the European Conference on Computer Vision, Milan, Italy."},{"key":"ref_42","doi-asserted-by":"crossref","unstructured":"Guia, J., Soares, V.G., and Bernardino, J. (2017, January 26\u201329). Graph databases: Neo4j analysis. Proceedings of the International Conference on Enterprise Information Systems, Porto, Portugal.","DOI":"10.5220\/0006356003510356"},{"key":"ref_43","first-page":"3","article-title":"Phi-2: The surprising power of small language models","volume":"1","author":"Javaheripi","year":"2023","journal-title":"Microsoft Res. Blog"},{"key":"ref_44","unstructured":"Bai, J., Bai, S., Chu, Y., Cui, Z., Dang, K., Deng, X., Fan, Y., Ge, W., Han, Y., and Huang, F. (2023). Qwen technical report. arXiv."}],"container-title":["Big Data and Cognitive Computing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2504-2289\/10\/1\/3\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,22]],"date-time":"2025-12-22T12:19:56Z","timestamp":1766405996000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2504-2289\/10\/1\/3"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,12,22]]},"references-count":44,"journal-issue":{"issue":"1","published-online":{"date-parts":[[2026,1]]}},"alternative-id":["bdcc10010003"],"URL":"https:\/\/doi.org\/10.3390\/bdcc10010003","relation":{},"ISSN":["2504-2289"],"issn-type":[{"value":"2504-2289","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,12,22]]}}}