Vol. 22, No. 3, pp. 255-262 (2026)
FATIGUE CRACK RECOGNITION AND QUANTIFICATION UNDER
COMPLEX BACKGROUND BASED ON REGION DETECTION
Yu-Hang Liu, Zhi-Yuan Yuan Zhou *, Bo-Hai Ji, Fei-Ya Rong and Tian Gao
School of Civil and Transportation Engineering, Hohai Univ., No.1, Xikang Rd., Nanjing 210000, China
*(Corresponding author: E-mail:This email address is being protected from spambots. You need JavaScript enabled to view it.)
Received: 25 March 2025; Revised: 23 July 2025; Accepted: 2 August 2025
DOI:10.18057/IJASC.2026.22.3.1
 View Article |
Export Citation: Plain Text | RIS | Endnote |
ABSTRACT
Accurate measurement of fatigue crack length is essential for routine maintenance of steel bridges, and automated technologies have already been applied in maintenance operations. However, conventional image processing strategies suffer from complex and multi-source interference in steel box girders. This study proposed a two-stage method combining region recognition and morphological processing to diminish noise in fatigue crack images, thereby improving the accuracy of crack recognition and length measurement. The object-recognition-former (OBFormer) model was built to detect and isolate the fatigue crack regions, leveraging the advanced feature extraction module and localization capabilities. Optimized Canny algorithm was utilized to suppress noise and extract the skeletal line of the fatigue crack from detected regions. The length of the crack was calculated by modifying the coordinate system of the crack plane and quantifying the number of pixels along the skeletal line. Experimental results showed that the OBFormer model effectively handled complex background interference and accurately detected the fatigue crack regions, with a mean average precision (mAP) of 91%. After morphological processing, the length of the crack skeleton was accurately calculated with a relative error of less than 5%. The proposed methodology provides a reliable and efficient solution for fatigue crack measurement, offering significant potential for real-world applications in structural health monitoring and bridge maintenance.
KEYWORDS
Steel box girder, Fatigue crack, Computer vision, Deep learning
REFERENCES
[1] Z.Q. Fu, B.H. Ji, D.D. Zhao, M.Y. Yang, Corrosion evaluation of steel material on steel box girder bridge by comprehensive scoring method, Materials Research Innovations 18 (2014) S2-840-S2-844. https://doi.org/10.1179/1432891714Z.000000000465.
[2] M.D. Sangid, The physics of fatigue crack initiation, International Journal of Fatigue 57 (2013) 58–72. https://doi.org/10.1016/j.ijfatigue.2012.10.009.
[3] K. Sun, X. Jiang, X. Qiang, Z. Lv, C. Fan, Equivalent Vehicle Model Based on Traffic Flow of Long-Span Steel Box Girder Bridge, J. Bridge Eng. 28 (2023) 04023061. https://doi.org/10.1061/JBENF2.BEENG-6086.
[4] Y. Wang, L. Chao, J. Chen, S. Jiang, Fatigue Crack Propagation Law of Corroded Steel Box Girders in Long Span Bridges, Computer Modeling in Engineering and Sciences 140 (2024) 201–227. https://doi.org/10.32604/cmes.2024.046129.
[5] X. Jiang, K. Sun, X. Qiang, D. Li, Q. Zhang, XFEM-based Fatigue Crack Propagation Analysis on Key Welded Connections of Orthotropic Steel Bridge Deck, Int J Steel Struct 23 (2023) 404–416. https://doi.org/10.1007/s13296-022-00701-3.
[6] C.-Y. Wang, A. Bochkovskiy, H.-Y.M. Liao, YOLOv7: Trainable bag-of-freebies sets new state-of-the-art for real-time object detectors, (2022). http://arxiv.org/abs/2207.02696 (accessed July 17, 2024).
[7] S. Teng, Z. Liu, G. Chen, L. Cheng, Concrete Crack Detection Based on Well-Known Feature Extractor Model and the YOLO_v2 Network, Applied Sciences 11 (2021) 813. https://doi.org/10.3390/app11020813.
[8] G. Ye, S. Li, M. Zhou, Y. Mao, J. Qu, T. Shi, Q. Jin, Pavement crack instance segmentation using YOLOv7-WMF with connected feature fusion, Automation in Construction 160 (2024) 105331. https://doi.org/10.1016/j.autcon.2024.105331.
[9] H. Zunair, A. Ben Hamza, Sharp U-Net: Depthwise convolutional network for biomedical image segmentation, Computers in Biology and Medicine 136 (2021) 104699. https://doi.org/10.1016/j.compbiomed.2021.104699.
[10] Y. Chu, X. Xiang, Y. Wang, B. Huang, Pavement Disease Detection through Improved YOLOv5s Neural Network, Computational Intelligence and Neuroscience 2022 (2022) 1–12. https://doi.org/10.1155/2022/1969511.
[11] G. Ye, J. Qu, J. Tao, W. Dai, Y. Mao, Q. Jin, Autonomous surface crack identification of concrete structures based on the YOLOv7 algorithm, Journal of Building Engineering 73 (2023) 106688. https://doi.org/10.1016/j.jobe.2023.106688.
[12] Z. Shi, N. Jin, D. Chen, D. Ai, A comparison study of semantic segmentation networks for crack detection in construction materials, Construction and Building Materials 414 (2024) 134950. https://doi.org/10.1016/j.conbuildmat.2024.134950.
[13] S. Qin, T. Qi, T. Deng, X. Huang, Image segmentation using Vision Transformer for tunnel defect assessment, Computer Aided Civil Eng (2024) mice.13181. https://doi.org/10.1111/mice.13181.
[14] R. Ali, J.H. Chuah, M.S.A. Talip, N. Mokhtar, M.A. Shoaib, Crack Segmentation Network using Additive Attention Gate—CSN-II, Engineering Applications of Artificial Intelligence 114 (2022) 105130. https://doi.org/10.1016/j.engappai.2022.105130.
[15] Y.-C. Chen, R.-T. Wu, A. Puranam, Multi-task deep learning for crack segmentation and quantification in RC structures, Automation in Construction 166 (2024) 105599. https://doi.org/10.1016/j.autcon.2024.105599.
[16] C. Dong, L. Li, J. Yan, Z. Zhang, H. Pan, F.N. Catbas, Pixel-Level Fatigue Crack Segmentation in Large-Scale Images of Steel Structures Using an Encoder–Decoder Network, Sensors 21 (2021) 4135. https://doi.org/10.3390/s21124135.
[17] Y. Xu, Y. Bao, J. Chen, W. Zuo, H. Li, Surface fatigue crack identification in steel box girder of bridges by a deep fusion convolutional neural network based on consumer-grade camera images, Structural Health Monitoring 18 (2019) 653–674. https://doi.org/10.1177/1475921718764873.
[18] E. Xie, W. Wang, Z. Yu, A. Anandkumar, J.M. Alvarez, P. Luo, SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers, Advances in Neural Information Processing Systems 34 (2021) 12077–12090.
[19] A. Vaswani, N. Shazeer, N. Parmar, J. Uszkoreit, L. Jones, A.N. Gomez, Ł. Kaiser, I. Polosukhin, Attention is All you Need, Advances in Neural Information Processing Systems 30 (2017).
[20] A. Dosovitskiy, L. Beyer, A. Kolesnikov, D. Weissenborn, X. Zhai, T. Unterthiner, M. Dehghani, M. Minderer, G. Heigold, S. Gelly, J. Uszkoreit, N. Houlsby, An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale, (2020). http://arxiv.org/abs/2010.11929 (accessed November 27, 2023).
[21] Z. Liu, Y. Lin, Y. Cao, H. Hu, Y. Wei, Z. Zhang, S. Lin, B. Guo, Swin Transformer: Hierarchical Vision Transformer using Shifted Windows, in: 2021 IEEE/CVF International Conference on Computer Vision (ICCV), IEEE, Montreal, QC, Canada, 2021: pp. 9992–10002. https://doi.org/10.1109/ICCV48922.2021.00986.
[22] I. Loshchilov, F. Hutter, SGDR: Stochastic Gradient Descent With Warm Restarts, in: International Conference on Learning Representations, 2017.
[23] I. Loshchilov, F. Hutter, Decoupled Weight Decay Regularization, arXiv Preprint arXiv:1711.05101 (2019). http://arxiv.org/abs/1711.05101 (accessed September 7, 2024).
[24] J. Canny, A Computational Approach to Edge Detection, IEEE Trans. Pattern Anal. Mach. Intell. PAMI-8 (1986) 679–698. https://doi.org/10.1109/TPAMI.1986.4767851.