Lightweight alternative machine learning model for automating concrete crack image classification

Abstract

In this study, the challenge of automating concrete crack image classification by developing a lightweight machine learning model that balances accuracy with computational efficiency was addressed. Traditional deep learning models, while accurate, suffer from high computational demands, limiting their practicality in on-site applications. This study's approach used the random forest (RF) classifier combined with histogram of oriented gradients (HOG) and local binary patterns (LBP) for feature extraction, offering a more feasible alternative for real-time structural health monitoring. Comparative analysis with the convolutional neural network (CNN) model highlights this model's significantly reduced size and inference times, with only a marginal compromise in accuracy. The results demonstrated that the RF models, particularly RF with LBP, are well-suited for integration into resource-constrained environments, paving the way for their deployment in portable, on-site diagnostic systems in civil engineering. This study contributed a novel perspective to the field, emphasizing the importance of efficient machine learning solutions in practical applications of structural health monitoring.

Cite

APA

Lee, S. M., Choi, H. S., Kim, C., & Kang, T. H.-K. (2025). Lightweight alternative machine learning model for automating concrete crack image classification. ACI Structural Journal, 122(5). https://www.proquest.com/docview/3249300050?pq-origsite=gscholar&fromopenview=true&sourcetype=Scholarly%20Journals

BibTeX

@article{lee2025lightweight,
  author  = {Lee, Sang Min and Choi, Hyun Su and Kim, Chul and Kang, Thomas H.-K.},
  title   = {Lightweight Alternative Machine Learning Model for Automating Concrete Crack Image Classification},
  journal = {ACI Structural Journal},
  volume  = {122},
  number  = {5},
  year    = {2025}
}