Automated system for damage analysis of urban infrastructure objects using computer vision models
DOI:
https://doi.org/10.30837/2522-9818.2026.2.121Keywords:
damage analysis; urban infrastructure; deep learning; computer vision models; instance segmentation; image set; intelligent systemAbstract
The current problem is to ensure the integration of multi-level data for effective monitoring and analysis of damage to buildings resulting from military operations in Ukraine, and to determine restoration priorities. The purpose of this work is to develop an intelligent system for automated analysis of damage to urban infrastructure objects, enabling effective processing of large volumes of images obtained using computer vision tools for further assessment of the level of destruction and decision-making regarding the priority of their restoration. The detailed reports generated by the system on each building’s condition significantly simplify the expert evaluation process, eliminating the need for specialists to be physically present on-site. Ranking damaged objects by their impact on the country’s vital activities ensures a more rational, well-balanced allocation of the resources required for infrastructure restoration. In addition, the system actively interacts with relevant state structures, local governments and specialists responsible for planning and organizing restoration work. Within the framework of such interaction, it is envisaged to obtain access to comprehensive databases with information on the architectural features of buildings, their historical and cultural value, as well as the use of existing technical documentation, which makes it possible to take into account all aspects when assessing the severity of damage and forming building restoration plans. Analysis of data obtained from specialised engineering surveys and studies using high-precision instrumental methods enables additional refinement of damage parameters and the creation of extended datasets for further training and validation of computer vision models. The paper investigates and justifies the feasibility of using instance segmentation with YOLOv11 and Mask R-CNN models, enabling not only damage detection but also accurate boundary and quantity determination, which is important for a comprehensive assessment of the scale of destruction and planning restoration measures. The results show that the proposed system works effectively when approaches to combat class imbalance are used. An important stage of the system is the automated formation of optimized restoration plans based on a comprehensive analysis of the identified damage. Such plans should take into account not only the technical parameters of the destruction, but also financial, material and personnel limitations. The proposed system is an effective tool for supporting decision-making at the state and regional levels on the restoration of damaged structures, contributing to the acceleration of the restoration processes of critical infrastructure and minimizing socio-economic losses associated with military destruction.Downloads
References
References
Gupta, R., Goodman, B., Patel, N., Hosfelt, R., Sajeev, S., Heim, E., Doshi, J., Curtis, K., Seferbekov, S., Alker, A. and Lee, S. (2019), "xBD: A Dataset for Assessing Building Damage from Satellite Imagery", arXiv, pp. 1–9. DOI: https://doi.org/10.48550/arXiv.1911.09296
Risso, M., Goffi, A., Motetti, B.A., Burrello, A., Bove, J.B., Macii, E., Poncino, M., Pagliari, D.J. and Maffeis, G. (2025), "Building Damage Assessment in Conflict Zones: A Deep Learning Approach Using Geospatial Sub-Meter Resolution Data", in Proceedings of the 2025 IEEE 6th International Conference on Image Processing, Applications and Systems (IPAS), pp. 1–6. DOI: https://doi.org/10.1109/IPAS63548.2025.10924534
Yuan, Q., Shen, H., Li, T., Li, Z., Li, S., Jiang, Y., Xu, H., Tan, W., Yang, Q., Wang, J., Gao, J. and Zhang, L. (2020), "Deep learning in environmental remote sensing: Achievements and challenges", Remote Sensing of Environment, 241, 111716. DOI: https://doi.org/10.1016/j.rse.2020.111716
Shevchenko, L. (2022), "Mass Housing in Ukraine in the Second Half of the 20th Century", Docomomo Journal, 67, pp. 72–78. DOI: https://doi.org/10.52200/docomomo.67.08
Teng, S., Chen, G., Liu, Z., Cheng, L. and Sun, X. (2021), "Multi-sensor and decision-level fusion-based structural damage detection using a one-dimensional convolutional neural network", Sensors, 21(12), 3950. DOI: https://doi.org/10.3390/s21123950
UN News (2025), "Ukraine: Post-War Reconstruction Set to Cost $524 Billion" [online]. Available from: https://news.un.org/en/story/2025/02/1160466 (Accessed: 2 April 2025).
UNDP Ukraine (n.d.) "In Ukraine, Machine-Learning Algorithms and Big Data Scans Used to Identify War-Damaged Infrastructure" [online]. Available from: https://www.undp.org/ukraine/blog/ukraine-machine-learning-algorithms-and-big-data-scans-used-identify-war-damaged-infrastructure (Accessed: 11 March 2025).
Geospatial World (n.d.) "Assessing Infra Damage in War-Torn Ukraine" [online]. Available from: https://geospatialworld.net/prime/technology-and-innovation/assessing-infra-damage-war-torn-ukraine/ (Accessed: 11 March 2025).
FlyPix (n.d.) "AI-Driven Building Damage Assessment: Innovations and Applications" [online]. Available from: https://flypix.ai/blog/building-damage-assessment/ (Accessed: 11 April 2025).
T2D2 (n.d.) "AI Damage Detection" [online]. Available from: https://t2d2.ai/damage-detector (Accessed: 11 March 2025).
WarUpClose (n.d.) "Memorialization of Destructions" [online]. Available from: https://war.city/about-us/ (Accessed: 23 April 2025).
Ukrainska Pravda (2023), "Rebuilding of Housing Destroyed by Russians Can Now Be Seen Online in 3D" [online]. Available from: https://www.pravda.com.ua/eng/news/2023/10/3/7422472/ (Accessed: 23 April 2025).
SCAN UA (n.d.) "Save Ukrainian Heritage" [online]. Available from: https://scanua.com/ (Accessed: 24 April 2025).
Kashtan, V. and Hnatushenko, V. (2023), "Automated building damage detection on digital imagery using machine learning", Naukovyi Visnyk Natsionalnoho Hirnychoho Universytetu, 6, pp. 134–140. DOI: https://doi.org/10.33271/nvngu/2023-6/134
Li, X., Caragea, D., Zhang, H. and Imran, M. (2018), "Localizing and quantifying damage in social media images", in Proceedings of the IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM), Barcelona, 28–31 August 2018. DOI: https://doi.org/10.1109/asonam.2018.8508298
Li, X., Caragea, D., Zhang, H. and Imran, M. (2019), "Localizing and quantifying infrastructure damage using class activation mapping approaches", Social Network Analysis and Mining, Vol. 9, No. 1, pp. 194–201. DOI: https://doi.org/10.1007/s13278-019-0588-4
Nia, K.R. and Mori, G. (2017), "Building damage assessment using deep learning and ground-level image data", in Proceedings of the 14th Conference on Computer and Robot Vision (CRV), Edmonton, 16–19 May 2017, pp. 95–102. DOI: https://doi.org/10.1109/CRV.2017.54
Klyshnia, A. (n.d.) "First Step Towards Circular Reconstruction: Damage Assessment and Material Mapping of Buildings in Ukraine Through Ground Imagery Processing" [online]. Available from: https://blog.iaac.net/first-step-towards-circular-reconstruction-damage-assessment-and-material-mapping-of-buildings-in-ukraine-through-ground-imagery-processing/ (Accessed: 24 April 2025).
Liu, C., Wang, Y., Zhang, J. and Li, H. (2022), "Real-time ground-level building damage detection based on lightweight and accurate YOLOv5 using terrestrial images", Remote Sensing, Vol. 14, No. 12, 2763. DOI: https://doi.org/10.3390/rs14122763
Bošnjak, A. and Jajac, N. (2023), "Determining priorities in infrastructure management using multicriteria decision analysis", Sustainability, Vol. 15, No. 20, 14953. DOI: https://doi.org/10.3390/su152014953
Mohammadnazari, Z., Mousapour, M., Alipour-Vaezi, M., Aghsami, A., Jolai, F. and Yazdani, M. (2022), "Prioritizing post-disaster reconstruction projects using an integrated multi-criteria decision-making approach: a case study", Buildings, Vol. 12, No. 2, 136. DOI: https://doi.org/10.3390/buildings12020136
Szeliski, R. (2022), Computer Vision: Algorithms and Applications, 2nd ed. Cham: Springer. DOI: https://doi.org/10.1007/978-3-030-34372-9
Liu, L., Ouyang, W., Wang, X., Fieguth, P., Chen, J., Liu, X. and Pietikäinen, M. (2020), "Deep learning for generic object detection: a survey", International Journal of Computer Vision, 128(2), pp. 261–318. DOI: https://doi.org/10.1007/s11263-019-01247-4
Krizhevsky, A., Sutskever, I. and Hinton, G.E. (2012), "ImageNet Classification with Deep Convolutional Neural Networks", Advances in Neural Information Processing Systems (NeurIPS), Vol. 25. DOI: https://doi.org/10.1145/3065386
He, K., Zhang, X., Ren, S. and Sun, J. (2016), "Deep Residual Learning for Image Recognition", in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 770–778. DOI: https://doi.org/10.1109/CVPR.2016.90
Manakitsa, N., Maraslidis, G.S., Moysis, L. and Fragulis, G.F. (2024), "A review of machine learning and deep learning for object detection, semantic segmentation, and human action recognition in machine and robotic vision", Technologies, Vol. 12, No. 2, 15. DOI: https://doi.org/10.3390/technologies12020015
Girshick, R. (2015), "Fast R-CNN", in Proceedings of the IEEE International Conference on Computer Vision (ICCV), pp. 1440–1448. DOI: https://doi.org/10.1109/ICCV.2015.169
Sultana, F., Sufian, A. and Dutta, P. (2020), "Evolution of image segmentation using deep convolutional neural networks: a survey", Knowledge-Based Systems, 201–202, 106062. DOI: https://doi.org/10.1016/j.knosys.2020.106062
Cheng, B., Schwing, A. and Kirillov, A. (2022), "Masked-attention Mask Transformer for Universal Image Segmentation (Mask2Former)", in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 1290–1299. DOI: https://doi.org/10.48550/arXiv.2112.01527
Wang, C.Y. et al. (2022) "YOLOv7: Trainable bag-of-freebies for real-time object detection", in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). DOI: https://doi.org/10.48550/arXiv.2207.02696
Zhao, Q. and Zhu, J. (2025), "An Improved YOLOv11 Architecture with Multi-Scale Attention and Spatial Fusion for Fine-Grained Residual Detection", Results in Engineering, 27, 107061. DOI: https://doi.org/10.1016/j.rineng.2025.107061
Huang, H., Zhao, S., Zhang, D. and Chen, J. (2020), "Deep learning-based instance segmentation of cracks from shield tunnel lining images", Structure and Infrastructure Engineering. DOI: https://doi.org/10.1080/15732479.2020.1838559
Kumar, T., Brennan, R., Mileo, A. and Bendechache, M. (2024), "Image data augmentation approaches: a comprehensive survey and future directions", IEEE Access, 12, pp. 187536–187571. DOI: https://doi.org/10.1109/ACCESS.2024.3470122
Johnson, J. and Khoshgoftaar, T. (2019), "Survey on deep learning with class imbalance", Journal of Big Data, 6(1), pp. 1–54. DOI: https://doi.org/10.1186/s40537-019-0192-5
Taherdoost, H. and Madanchian, M. (2023), "Multi-Criteria Decision Making (MCDM) methods and concepts", Encyclopedia, 3(1), pp. 77–87. DOI: https://doi.org/10.3390/encyclopedia3010006
Madanchian, M. and Taherdoost, H. (2023), "A comprehensive guide to the TOPSIS method for multi-criteria decision making", Sustainable Social Development, 1(1), pp. 1–6. DOI: https://doi.org/10.54517/ssd.v1i1.2220
Jagerman, R., Oosterhuis, H. and de Rijke, M. (2022), "To model or to intervene: A comparison of counterfactual and online learning to rank from user interactions", in Proceedings of the ACM International Conference on Web Search and Data Mining. DOI: https://doi.org/10.1145/3331184.3331269
Aftab, S. and Ramampiaro, H. (2024), "Improving top-N recommendations using batch approximation for weighted pair-wise loss", Machine Learning with Applications, 15, 100520. DOI: https://doi.org/10.1016/j.mlwa.2023.100520
Buda, M., Maki, A. and Mazurowski, M.A. (2018), "A systematic study of the class imbalance problem in convolutional neural networks", Neural Networks, 106, pp. 249–259. DOI: https://doi.org/10.1016/j.neunet.2018.07.011
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Our journal abides by the Creative Commons copyright rights and permissions for open access journals.
Authors who publish with this journal agree to the following terms:
-
Authors hold the copyright without restrictions and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License (CC BY-NC-SA 4.0) that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal.
-
Authors are able to enter into separate, additional contractual arrangements for the non-commercial and non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
-
Authors are permitted and encouraged to post their published work online (e.g., in institutional repositories or on their website) as it can lead to productive exchanges, as well as earlier and greater citation of published work.
