Pavement Crack Detection and Classification Using Deep Learning Techniques
Keywords:
Pavement crack analysis, crack categorization, deep learning, ResNet, attention mechanismAbstract
Accurate detection and analysis of pavement cracks are essential components of intelligent road maintenance systems. The primary challenges in this area stem from the irregular shapes, complex topologies, and texture-like noise often associated with cracks. To address these issues, this study introduces a comprehensive and robust automated framework capable of identifying crack types and evaluating their severity levels. A specialized crack detection network is designed, incorporating a multi-scale dilated convolution module to effectively extract contextual information about cracks. An attention mechanism is integrated to enhance the refinement of high-level features. These enriched features are then fused through an upsampling module, enabling the generation of detailed detection outputs. Following detection, a novel classification approach is applied to categorize the cracks into four types—transverse, longitudinal, block, and alligator. Additionally, the severity of the cracks is quantified by measuring the average width and spacing between crack branches. This integrated system offers an effective solution for real-time pavement assessment and management.
Downloads
References
Federal Highway Administration, "Distress Identification Manual for the Long-Term Pavement Performance Program," FHWA-RD-03-031, 2003. [Online]. Available: https://highways.dot.gov/sites/fhwa.dot.gov/files/FHWA-RD-03-031.pdf
Y. Yang et al., "Efficient pavement crack detection and classification," EURASIP Journal on Image and Video Processing, vol. 2017, no. 1, pp. 1-11, 2017. [Online]. Available: https://jivp-eurasipjournals.springeropen.com/articles/10.1186/s13640-017-0187-0
J. Zhang et al., "A method of hybrid dilated and global convolution networks for pavement crack detection," Multimedia Tools and Applications, vol. 81, no. 2, pp. 1-19, 2022. [Online]. Available: https://link.springer.com/article/10.1007/s00530-024-01408-7
G. Zhu et al., "RHA-Net: An Encoder-Decoder Network with Residual Blocks and Hybrid Attention Mechanisms for Pavement Crack Segmentation," arXiv preprint arXiv:2207.14166, 2022. [Online]. Available: https://arxiv.org/abs/2207.14166
M. Rababaah et al., "Asphalt Pavement Crack Classification: A Comparison of GA, MLP, and SOM," International Journal of Computer Applications, vol. 78, no. 1, pp. 1-6, 2013. [Online]. Available: https://www.cs.iusb.edu/~isl/jim_papers/78-rababaah.pdf
Federal Highway Administration, "Distress Identification Manual," FHWA-RD-03-031, 2003. [Online]. Available: https://highways.dot.gov/sites/fhwa.dot.gov/files/docs/research/long-term-pavement-performance/products/1401/distress-identification-manual-13092.pdf
Fan, R., Ai, Y., Dahnoun, N., & McAllister, K. (2018). “Road surface condition monitoring using smart sensing and deep learning”. IEEE Transactions on Industrial Informatics, 14(9), 4032–4041. https://doi.org/10.1109/TII.2018.2803205
Yang, F., Zhang, L., & Yu, S. (2019). “Pavement crack detection and segmentation based on deep learning with attention mechanism”. In Proceedings of the 2019 IEEE International Conference on Image Processing (ICIP), 4754–4758. https://doi.org/10.1109/ICIP.2019.8803521
H. Oliveira and P. L. Correia, “Automatic road crack detection and characterization,” IEEE Transactions on Intelligent Transportation Systems, vol. 14, no. 1, pp. 155–168, 2013.
Q. Zou, Y. Cao, Q. Li, Q. Mao, and S. Wang, “CrackTree: Automatic crack detection from pavement images,” Pattern Recognition Letters, vol. 33, no. 3, pp. 227–238, 2012.
S. Chambon and J. P. Moliard, “Automatic road pavement assessment with image processing: Review and comparison,” International Journal of Geophysics, vol. 2011, Article ID 989354, 2011.
L. Zhang, F. Yang, Y. D. Zhang, and Y. J. Zhu, “Road crack detection using deep convolutional neural network,” in Proc. IEEE Int. Conf. Image Process. (ICIP), 2016, pp. 3708–3712.
W. Shi, B. Zhang, and H. Zhang, “Automatic road crack detection using D-LinkNet,” Sensors, vol. 20, no. 18, p. 5105, 2020.
Z. Chen, Y. Liu, Y. Wang, and Y. Yu, “RAUNet: Residual Attention U-Net for road crack segmentation,” Automation in Construction, vol. 128, p. 103775, 2021.
M. R. Jahanshahi, J. F. Masri, and F. Padgett, “An innovative methodology for detection and quantification of cracks through incorporation of depth perception,” Machine Vision and Applications, vol. 24, pp. 227–241, 2013.
Y. Li, L. Zhao, and Z. Wang, “Real-time crack detection and classification using convolutional and recurrent neural networks,” Structural Health Monitoring, vol. 20, no. 1, pp. 46–58, 2021.
K. Gopalakrishnan, “Deep learning in data-driven pavement image analysis and modeling: Recent advances and future directions,” Journal of Transportation Engineering, Part B: Pavements, vol. 144, no. 1, p. 04017061, 2018.
X. Zhao, J. Zhang, and H. Su, “Crack detection using unmanned aerial vehicle (UAV) based imagery and deep learning,” Automation in Construction, vol. 119, p. 103312, 2020.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 International Journal of Scientific Research in Science, Engineering and Technology
This work is licensed under a Creative Commons Attribution 4.0 International License.
