Predictive maintenance in infrastructure: Utilizing 3D point clouds for efficient damage detection
DOI:
https://doi.org/10.71911/cii-p3-nt-2025223Keywords:
tree roots, predictive maintenance, damage detection, 3D point cloudAbstract
Growing urbanization is driving the demand for infrastructure such as parking lots, roads, and bicycle lanes. While green spaces and trees are often integrated into these development projects to mitigate negative climate impacts, they can cause root-related damage that poses safety risks and requires costly monitoring. Public road networks are typically inspected with advanced but expensive surveillance vehicles that are too costly for private applications, leaving private infrastructure such as parking lots, private roads, and storage areas without comparable solutions. Thus, this paper presents a methodology for detecting and classifying damage areas in 3D point clouds of parking lots, distinguishing root-related damage from construction joints using a combination of deep learning and classical statistics. The approach is evaluated on data from Vienna International Airport and validated against manually labeled ground truth data. Results show that accurate localization and classification of damage is feasible using only a single laser scanner, providing a cost-effective alternative to conventional monitoring. Moreover, the method facilitates predictive maintenance by automatically detecting damage and enabling integration into Building Information Modeling software.
References
[1] United Nations Department of Economic and Social Affairs, “2018 Revision of World Urbanization Prospects,” United Nations, Department of Economic and Social Affairs, 2019
[2] R. Scalenghe and F. A. Marsan, "The anthropogenic sealing of soils in urban areas," ,Landscape and Urban Planning, vol. 90, no. 1-2, p. 1–10, 2009, doi: https://doi.org/10.1016/j.landurbplan.2008.10.011
[3] Z. Ring, D. Damyanovic and F. Reinwald, "Green and open space factor Vienna: A steering and evaluation tool for urban green infrastructure," Urban Forestry & Urban Greening, vol. 62, p. 127131, 2021, doi: https://doi.org/10.1016/j.ufug.2021.127131
[4] S. Jaafari, A. Shabani, M. Moeinaddini, A. Danehkar and Y. Sakieh, "Applying landscape metrics and structural equation modeling to predict the effect of urban green space on air pollution and respiratory mortality in Tehran," Environmental Monitoring and Assessment, vol. 192, no. 412, 15 p., 2020, doi: https://doi.org/10.1007/s10661-020-08377-0
[5] L. Sturiale and A. Scuderi, "the role of green infrastructures in urban planning for climate change adaptation," Climate, vol. 7, no. 10, p. 119, 2019, doi: https://doi.org/10.3390/cli7100119
[6] H. Maeda, Y. Sekimoto, T. Seto, T. Kashiyama and H. Omata, "Road damage detection and classification using deep neural networks with smartphone images," Computer-Aided Civil and Infrastructure Engineering, vol. 33, no. 12, p. 1127–1141, 2018, doi: https://doi.org/10.1111/mice.12387
[7] M. E. Mohammadi, D. P. Watson and R. L. Wood, "Deep learning-based damage detection from aerial SfM point clouds," Drones, vol. 3, no. 3, p. 68, 2019, doi: https://doi.org/10.3390/drones3030068
[8] Z. Feng, A. El Issaoui, M. Lehtomäki, M. Ingman, H. Kaartinen, A. Kukko, et al., "Pavement distress detection using terrestrial laser scanning point clouds – Accuracy evaluation and algorithm comparison," ISPRS Open Journal of Photogrammetry and Remote Sensing, vol. 3, p. 100010, 2022, doi: https://doi.org/10.1016/j.ophoto.2021.100010
[9] C. Petschnigg, M. Spitzner, L. Weitzendorf and J. Pilz, "From a point cloud to a simulation model—Bayesian segmentation and entropy based uncertainty estimation for 3D modelling," Entropy, vol. 23, no. 3, p. 301, 2021, doi: https://doi.org/10.3390/e23030301
[10] CloudCompare, version 2.14, CloudCompare, available: https://www.cloudcompare.org/release/, 2025
[11] R. Q. Charles, H. Su, M. Kaichun, and L. J. Guibas, "PointNet: deep learning on point sets for 3D classification and segmentation," in 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 77-85, 2017, doi: https://doi.org/10.1109/CVPR.2017.16
[12] C. Choy, J. Gwak, and S. Savarese, "4D spatio-temporal convnets: Minkowski convolutional neural networks," in 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pp. 3070-3079, 2019, doi: https://doi.org/10.1109/CVPR.2019.00319
[13] J. Deng, W. Zhou, Y. Zhang and H. Li, “From multi-view to hollow-3D: hallucinated hollow-3D R-CNN for 3D object detection,” IEEE Transactions on Circuits and Systems for Video Technology, vol. 31, no. 12, p. 4722–4734, 2021, doi: https://doi.org/10.1109/tcsvt.2021.3100848
[14] A. Paszke, S. Gross, F. Massa, A. Lerer, J. Bradbury, G. Chanan, et al., “PyTorch: An imperative style, high-performance deep learning library,” in 33rd Conference on Neural Information Processing Systems, 12 p., 2019, doi: https://doi.org/10.48550/arXiv.1912.01703
Published
Issue
Section
License
Copyright (c) 2025 Carinthia II / Part 3 - Carinthia Nature Tech
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
