ตัวแบบการใช้โครงข่ายประสาทเทียมแบบสังวัตนาการเพื่อตรวจหารอยร้าวคอนกรีต

Prawat Suwanpraditkul; Wipawan Buathong; Pita Jarupunphol

Authors

Prawat Suwanpraditkul Program in Digital Technology, Faculty of Science and Terchnology, Phuket Rajabhat University
Wipawan Buathong Program in Digital Technology, Faculty of Science and Terchnology, Phuket Rajabhat University
Pita Jarupunphol Program in Digital Technology, Faculty of Science and Terchnology, Phuket Rajabhat University

Keywords:

machine learning, concrete crack detection, convolutional neural network

Abstract

This research aims to develop a model for using artificial intelligence using convolutional neural networks (CNN) to automatically detect concrete cracks in construction work, reducing the need for manual inspection. A diverse dataset comprising various concrete crack scenarios is employed to train and test, ensuring robustness in crack detection across different contexts. In addition to improving accuracy and consistency, this machine learning-based approach enhances safety by enabling the inspection of difficult-to-reach or hazardous areas. Furthermore, it enhances scalability and efficiency by rapidly assessing large or multiple sites. This approach allows for early intervention, which can prevent potential structural failures, minimise repair costs, and extend the lifespan of concrete infrastructures. The outcomes indicate that the model is highly effective, accurately classifying 39,482 out of 40,000 instances, which equates to an accuracy rate of 98.705%. Moreover, the weighted average metrics, including True Positive Rate, False Positive Rate, Precision, Recall, and F-Measure, all closely align with this high degree of accuracy, with almost all metrics at 0.987, demonstrating that the CNN model performance in accuracy and efficiency. Meanwhile, the results from the 70/30 Percentage Split method are in the same direction, representing the model's high performance. This yielded 11,829 (98.575%) correct instance predictions and only 171 (1.425%) incorrect instances.

References

Cement, G. F. P. (12 September 2019). The Importance of Concrete in Construction Projects. GFP Cement Contractors LLC. https://gfpcement.com/the-importance-of-concrete-in-construction-projects/

Ding, W., Yang, H., Yu, K., and Shu, J. (2023). Crack detection and quantification for concrete structures using UAV and transformer. Automation in Construction, 152, 104929. https://doi.org/10.1016/j.autcon.2023.104929

Golding, V. P., Gharineiat, Z., Munawar, H. S., and Ullah, F. (2022). Crack Detection in Concrete Structures Using Deep Learning. Sustainability, 14(13), 8117. https://doi.org/10.3390/su14138117

Imran, H., Al-Abdaly, N. M., Shamsa, M. H., Shatnawi, A., Ibrahim, M., and Ostrowski, K. A. (2022). Development of Prediction Model to Predict the Compressive Strength of Eco-Friendly Concrete Using Multivariate Polynomial Regression Combined with Stepwise Method. Materials, 15(317), 317. https://doi.org/10.3390/ma15010317

Kaggle. (n.d.). Concrete Crack Images for Classification. Kaggle. https://www.kaggle.com/datasets/arnavr10880/concrete-crack-images-for-classification.

Li, Z., Yoon, J., Zhang, R., Rajabipour, F., Iii, W. V. S., Dabo, I., and Radlinska, A. (2022). Machine learning in concrete science: Applications, challenges, and best practices. Npj Computational Materials, 8(1), 1–17. https://doi.org/10.1038/s41524-022-00810-x

Rubio, J. de J., Garcia, D., Rosas, F. J., Hernandez, M. A., Pacheco, J., and Zacarias, A. (2024). Stable convolutional neural network for economy applications. Engineering Applications of Artificial Intelligence, 132, 107998. https://doi.org/10.1016/j.engappai.2024.107998

Sami, B. H. Z., Sami, B. F. Z., Kumar, P., Ahmed, A. N., Amieghemen, G. E., Sherif, M. M., and El-Shafie, A. (2023). Feasibility analysis for predicting the compressive and tensile strength of concrete using machine learning algorithms. Case Studies in Construction Materials, 18, e01893. https://doi.org/10.1016/j.cscm.2023.e01893

Why concrete testing is important in engineering & construction. (2 november 22022). Douglas Partners. https://www.douglaspartners.com.au/news/the-importance-of-concrete-testing.

Witten, I. H., Frank, E., Hall, M. A., and Pal, C. J. (2016). Data Mining: Practical Machine Learning Tools and Techniques. In Data Mining: Practical Machine Learning Tools and Techniques. https://doi.org/10.1016/c2009-0-19715-5.

Yasin, M., Sarıgül, M., and Avci, M. (2024). Logarithmic Learning Differential Convolutional Neural Network. Neural Networks, 172, 106114. https://doi.org/10.1016/j.neunet.2024.106114

Yokoyama, S., and Matsumoto, T. (2017). Development of an Automatic Detector of Cracks in Concrete Using Machine Learning. Procedia Engineering, 171, 1250–1255. https://doi.org/10.1016/j.proeng.2017.01.418