Use of Machine Learning Algorithms for Assessing the Degree of Damage to Residential Infrastructure
DOI:
https://doi.org/10.52575/2687-0932-2025-52-1-156-167Keywords:
machine learning, infrastructure damage assessment, convolutional neural networks (CNN), YOLOv5, ResNetAbstract
In the modern world, the application of machine learning (ML) algorithms has become increasingly critical across various domains, particularly in assessing damage to residential infrastructure following emergencies such as natural disasters, military conflicts, or human-induced incidents. This study explores the integration of convolutional neural networks (CNNs), recurrent neural networks (RNNs), and hybrid architectures (CNN-RNN) to automate damage evaluation, enabling rapid and precise crisis response. By analyzing aerial imagery, satellite data, and crowdsourced visual content, ML models like ResNet50, ResNet101, YOLOv5, and VGG19 demonstrate significant potential in detecting structural damage, classifying severity levels, and prioritizing recovery efforts. Experimental results from benchmark datasets (e.g., post-explosion imagery from Beirut, earthquake-damaged buildings in Wenchuan) reveal that YOLOv5-based architectures achieve up to 93.43 % mean average precision (mAP) in real-time object detection, while ResNet50 attains 95.92 % accuracy in multi-class damage classification. However, challenges persist in recognizing fine-grained defects (e.g., cracks, spalls) due to resolution limitations and occlusions. The study highlights the advantages of attention mechanisms (e.g., CBAM, Ghost bottlenecks) and data augmentation techniques in improving model robustness. Comparative analysis of traditional manual inspections versus ML-driven approaches underscores the latter’s efficiency in resource allocation, accelerated reconstruction timelines, and enhanced safety protocols.
Downloads
References
Chachra G., Kong Q., Huang J., Korlakunta S., Grannen J., Robson A., Allen R.M. 2022. Detecting damaged buildings using real-time crowdsourced images and transfer learning. Scientific reports, 12(1), 8968. DOI: 10.1038/s41598-022-12965-0.
Chicco D., Jurman G. 2020. The advantages of the Matthews correlation coefficient (MCC) over F1 score and accuracy in binary classification evaluation. BMC genomics, 21, 1–13. DOI: 10.1186/s12864-019-6413-7.
Creswell A., White T., Dumoulin V., Arulkumaran K., Sengupta B., Bharath A.A. 2018. Generative adversarial networks: An overview. IEEE signal processing magazine, 35(1), 53–65. DOI: 10.1109/MSP.2017.2765202.
Han K., Wang Y., Tian Q., Guo J., Xu C. 2020. Ghostnet: More features from cheap operations. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition (pp. 1580–1589). DOI: 10.48550/arXiv.1911.11907.
Hansen J.G., de Figueiredo R.P. 2024. Active Object Detection and Tracking Using Gimbal Mechanisms for Autonomous Drone Applications. Drones, 8(2), 55. DOI: 10.3390/drones8020055.
Haryono A., Jati G., Jatmiko W. 2024. Oriented object detection in satellite images using convolutional neural network based on ResNeXt. ETRI Journal, 46(2), 307–322. DOI: 10.4218/etrij.2022-0446.
Henderson P., Ferrari V. 2017. End-to-end training of object class detectors for mean average precision. In Computer Vision–ACCV 2016: 13th Asian Conference on Computer Vision, Taipei, Taiwan, November 20-24, 2016, Revised Selected Papers, Part V 13 (pp. 198–213). Springer International Publishing. DOI: 10.1007/978-3-319-54193-8_13.
Hu S., Wang P., Hoare C., O’Donnell J. 2022. Building occupancy detection and localization using CCTV camera and deep learning. IEEE Internet of Things Journal, 10(1), 597–608. DOI: 10.1109/JIOT.2022.3201877.
Indumathi, C. P., Santhoshsivan, V., Selvakumar, R. (2024). ResNet and ResNeSt-Based Deep-Learning. In Digital Geography: Proceedings of the International Conference on Internet and Modern Society (IMS 2023) (p. 215). Springer Nature. DOI: 10.1007/978-3-031-67762-5_17.
Jiang P., Ergu D., Liu F., Cai Y. Ma B. 2022. A Review of Yolo algorithm developments. Procedia computer science, 199, 1066–1073. DOI: 10.1016/j.procs.2022.01.135.
Kallas J., Napolitano R. 2023. Automated large-scale damage detection on historic buildings in post-disaster areas using image segmentation. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 48, 797–804. DOI: 10.5194/isprs-archives-XLVIII-M-2-2023-797-2023.
Karacı A. 2022. VGGCOV19-NET: automatic detection of COVID-19 cases from X-ray images using modified VGG19 CNN architecture and YOLO algorithm. Neural Computing and Applications, 34(10), 8253–8274.
Kaur R., Singh S. 2023. A comprehensive review of object detection with deep learning. Digital Signal Processing, 132, 103812. DOI: 10.1016/j.dsp.2022.103812.
Khan S., Naseer M., Hayat M., Zamir S. W., Khan F. S., Shah, M. 2022. Transformers in vision: A survey. ACM computing surveys (CSUR), 54(10s), 1–41. DOI: 10.1145/3505244.
Krizhevsky A., Sutskever I., Hinton G.E. 2012. Imagenet classification with deep convolutional neural networks. Advances in neural information processing systems, 25. DOI: 10.1145/3065386.
Krizhevsky, A., Sutskever, I., Hinton, G.E. 2017. ImageNet classification with deep convolutional neural networks. Communications of the ACM, 60(6), 84–90. DOI: 10.1145/3065386.
Li X., Caragea D., Zhang H., Imran M. 2018. Localizing and quantifying damage in social media images. In 2018 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM) (pp. 194–201). IEEE. DOI: 10.1109/ASONAM.2018.8508298.
Lin C.L., Wu K.C. 2023. Development of revised ResNet-50 for diabetic retinopathy detection. BMC bioinformatics, 24(1), 157. DOI: 10.1186/s12859-023-05293-1.
Liu C., Sui H., Wang J., Ni, Z., Ge L. 2022. Real-time ground-level building damage detection based on lightweight and accurate YOLOv5 using terrestrial images. Remote Sensing, 14(12), 2763. DOI: 10.3390/rs14122763.
Mikołajczyk A., Grochowski M. 2018. Data augmentation for improving deep learning in image classification problem. In 2018 international interdisciplinary PhD workshop (IIPhDW) (pp. 117–122). IEEE. DOI: 10.1109/IIPHDW.2018.8388338.
Selvaraju R.R., Cogswell M., Das A., Vedantam, R., Parikh, D., Batra, D. 2017. Grad-cam: Visual explanations from deep networks via gradient-based localization. In Proceedings of the IEEE international conference on computer vision (pp. 618–626). DOI: 10.1007/s00521-022-06918-x.
Van Etten, A. 2018. You only look twice: Rapid multi-scale object detection in satellite imagery. DOI: 10.48550/arXiv.1805.09512.
Woo S., Park J., Lee J.Y., Kweon I.S. 2018. Cbam: Convolutional block attention module. In Proceedings of the European conference on computer vision (ECCV) (pp. 3-19). DOI: 10.48550/arXiv.1807.06521
Zhang Q. 2022. A novel ResNet101 model based on dense dilated convolution for image classification. SN Applied Sciences, 4, 1–13. DOI: 10.1007/s42452-021-04897-7.
Zhao D., Shao F., Liu Q., Yang L., Zhang H., Zhang, Z. 2024. A Small Object Detection Method for Drone-Captured Images Based on Improved YOLOv7. Remote Sensing, 16(6), 1002. DOI: 10.3390/rs16061002.
Zhu T., Chen J., Zhu R., Gupta G. 2023. StyleGAN3: generative networks for improving the equivariance of translation and rotation. arXiv preprint arXiv:2307.03898. DOI: 10.48550/arXiv.2307.03898.
Abstract views: 12
Share
Published
How to Cite
Issue
Section
Copyright (c) 2025 Economics. Information Technologies
This work is licensed under a Creative Commons Attribution 4.0 International License.
