Cracks Detection in Concrete Surfaces using Machine Learning
DOI:
https://doi.org/10.5433/1679-0375.2025.v46.51427Keywords:
infrastructure maintenance, computer vision, texture descriptors, automated monitoringAbstract
In this study, Machine Learning techniques were used to detect cracks in concrete surfaces. Texture descriptors were extracted from a set of images of cracked and non-cracked concrete surfaces using Gray Level Co-occurrence Matrix (GLCM). Texture descriptors features were used to train four machine learning models: Logistic Regression, Artificial Neural Networks, Random Forest, and eXtreme Gradient Boosting (XGBoost). k-fold cross validation was used to evaluate the performance of the models as well as metrics such as accuracy, sensitivity, precision, F1-score - the harmonic mean between precision and sensitivity - and the Area Under the ROC Curve (AUC). The results indicate satisfactory performance of all models, with emphasis on the XGBoost model, which achieved 99.65% accuracy and 99.70% sensitivity.
Downloads
References
Arivabene, A. C. (2015). Patologias em estruturas de concreto armado: Estudo de caso. Revista Especialize On-line IPOG, 1(10), 1-22.
Associação Brasileira de Normas Técnicas. (2010). NBR 9575: Impermeabilização - Seleção e projeto. ABNT.
Associação Brasileira de Normas Técnicas. (2014). NBR 6118: Projeto de estruturas de concreto - Procedimento. ABNT.
Chen, K., Yadav, A., Khan, A., Meng, Y., & Zhu, K. (2019). Improved crack detection and recognition based on convolutional neural network. Modelling and Simulation in Engineering, 2019 1–8. https://doi.org/10.1155/2019/8796743
Diniz, J. de C. N., Paiva, A. C., Braz, Geraldo, Jr., Almeida, J. D. S., Silva, A. C., Cunha, A. M. T. S., & Cunha, S. C. A. P. S. (2023). A method for detecting pathologies in concrete structures using deep neural networks. Applied Sciences, 13(9), 5763. https://doi.org/10.3390/app13095763
Haralick, R. M., Shanmugam, K., & Dinstein, I. H. (1973). Textural features for image classification. IEEE Transactions on systems, man, and cybernetics, (6), 610-621. 10.1109/TSMC.1973.4309314
Kim, J. J., Kim, A.-R., & Lee, S.-W. (2020). Artificial neural network-based automated crack detection and analysis for the inspection of concrete structures. Applied Sciences, 10(22), 8105. https://doi.org/10.3390/app10228105
Mansour, I. R., & Thomson, R. M. (2023). Haralick texture feature analysis for characterization of specific energy and absorbed dose distributions across cellular to patient length scales. Physics in Medicine & Biology, 68(9),1-12. https://doi.org/10.1088/1361-6560/acc7f1
Nascimento, E. R. dos S., & Fontes, M. D. S. (2021). Patologias das estruturas de concreto armado. Revista FATEC de Tecnologia e Ciências, 6(1), 1-28. https://revista.fatecba.edu.br/index.php/rftc/issue/view/rftcv6n12021
Özgenel, Ç. F. (2019). Concrete crack images for classification. Mendeley Data, 2. https://doi.org/10.17632/5y9wdsg2zt.2
Pacifici, F., Chini, M., & Emery, W. J. (2009). A neural network approach using multiscale textural metrics from very high-resolution panchromatic imagery for urban land use classification. Remote Sensing of Environment, 113(6), 1276–1292. https://doi.org/10.1016/j.rse.2009.02.014
R Core Team. (2024). R: A language and environment for statistical computing. R Foundation for Statistical Computing.
Tou, J., Tou, P., Lau, P. Y., & Tay, Y. H. (2007). Computer vision-based wood recognition system. In Thai Embedded Systems Association, Proceedings of the International Workshop on Advanced Image Technology [Proceedings]. International Workshop on Advance Image Technology, Bangkok, Thailand.
Vrbik, I., Van Nest, S. J., Meksiarun, P., Loeppky, J., Brolo, A., Lum, J. J., & Jirasek, A. (2019). Haralick texture feature analysis for quantifying radiation response heterogeneity in murine models observed using Raman spectroscopic mapping. PLoS One, 14(2), e0212225. 10.1371/journal.pone.0212225
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 Antonio Mendes Magalhães Júnior, Ícaro Viterbre Debique Sousa, Heron Viterbre Debique Sousa, Thelma Sáfadi, Paulo Henrique Sales Guimarães
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
The Copyright Declaration for articles published in this journal is the author’s right. Since manuscripts are published in an open access Journal, they are free to use, with their own attributions, in educational and non-commercial applications. The Journal has the right to make, in the original document, changes regarding linguistic norms, orthography, and grammar, with the purpose of ensuring the standard norms of the language and the credibility of the Journal. It will, however, respect the writing style of the authors. When necessary, conceptual changes, corrections, or suggestions will be forwarded to the authors. In such cases, the manuscript shall be subjected to a new evaluation after revision. Responsibility for the opinions expressed in the manuscripts lies entirely with the authors.
This journal is licensed with a license Creative Commons Attribution-NonCommercial 4.0 International.
