A Novel Approach to Multi-Class Crack Classification: ResNet50 Enhanced with Squeeze-and-Excitation Blocks
Keywords:
Concrete structures, Classification accuracy, Crack classification , Feature Extraction , Squeeze and ExcitationAbstract
This study focuses on introducing a novel multi-class crack classification framework for concrete structures by integrating Squeeze-and-Excitation (SE) blocks into the ResNet50 architecture. Unlike conventional models, which lack feature recalibration capabilities, the embedded SE block adaptively enhances channel-wise feature importance, significantly improving crack detection performance under complex conditions such as lighting variations and background noise. The proposed model is trained on a comprehensive dataset of 7,147 images, derived from the Özgenel dataset, the Concrete Crack Conglomerate Dataset, and additional images collected from buildings at Universiti Teknologi Malaysia (UTM). Experimental results demonstrate that ResNet50-SE achieves a remarkable accuracy of 99.88%, outperforming the standard ResNet50 (93.25%) and surpassing existing multi-class crack classification models. Furthermore, the optimized architecture ensures computational efficiency, making it well-suited for real-time structural health monitoring applications. These findings validate the effectiveness of SE-enhanced architectures in improving feature extraction and classification accuracy, marking a significant advancement in automated crack detection.References
[1] D. Kang, S. S. Benipal, D. L. Gopal and Y.-J. Cha, Hybrid pixel-level concrete crack segmentation and quantification across complex backgrounds using deep learning, Automation in Construction, 118, 2020, 103291.
[2] B. Kim and S. Cho, Automated vision-based detection of cracks on concrete surfaces using a deep learning technique, Sensors, 18(10), 2018, 3452.
[3] F. Fang, L. Li, Y. Gu, H. Zhu and J. H. Lim, A novel hybrid approach for crack detection, Pattern Recognition, 107, 2020, 107474.
[4] S. Dorafshan, R. J. Thomas and M. Maguire, Comparison of deep convolutional neural networks and edge detectors for image-based crack detection in concrete, Construction and Building Materials, 186, 2018, 1031-1045.
[5] Y. Liu, J. K. W. Yeoh and D. K. H. Chua, Deep learning–based enhancement of motion blurred UAV concrete crack images, Journal of Computing in Civil Engineering, 34(5), 2020, 04020028.
[6] M. Mazni, A. R. Husain, M. I. Shapiai and I. S. Ibrahim, Crack recognition in masonry structures : CNN models with limited data sets, ELEKTRIKA: Journal of Electrical Engineering, 23(1), 2024, 133-140.
[7] D. Liang, X. F. Zhou, S. Wang and C. J. Liu, Research on concrete cracks recognition based on dual convolutional neural network, KSCE Journal of Civil Engineering, 23(7), 2019, 3066-3074.
[8] H. Xu, X. Su, H. Xu and H. Li, Autonomous bridge crack detection using deep convolutional neural networks, Proceedings of the 3rd International Conference on Computer Engineering, Information Science & Application Technology, 2019.
[9] S. Li and L. Sun, Detectability of bridge-structural damage based on fiber-optic sensing through deep-convolutional neural networks, Journal of Bridge Engineering, 25(4), 2020, 04020012.
[10] E. McLaughlin, N. Charron and S. Narasimhan, Automated defect quantification in concrete bridges using robotics and deep learning, Journal of Computing in Civil Engineering, 34(5), 2020, 04020029.
[11] T. S. Tran, V. P. Tran, H. J. Lee, J. M. Flores and V. P. Le, A two-step sequential automated crack detection and severity classification process for asphalt pavements, International Journal of Pavement Engineering, 23(6), 2022, 2019-2033.
[12] Q. Yang, W. Shi, J. Chen and W. Lin, Deep convolution neural network-based transfer learning method for civil infrastructure crack detection, Automation in Construction, 116, 2020, 103199.
[13] T. U. Ahmed, M. Shahadat Hossain, M. J. Alam and K. Andersson, An integrated CNN-RNN framework to assess road crack, 22nd International Conference on Computer and Information Technology (ICCIT 2019), Dhaka, Bangladesh, 2019, 18-20.
[14] L. Ali, F. Alnajjar, W. Khan, M. A. Serhani and H. A. Jassmi, Bibliometric analysis and review of deep learning-based crack detection literature published between 2010 and 2022, Buildings, 12(4), 432.
[15] M. Mazni, A. R. Husain, M. I. Shapiai, I. S. Ibrahim and D. W. Anggara, Identification of concrete cracks using deep learning models: A systematic review, Applications of Modelling and Simulation, 8(1), 2024, 1-25.
[16] M. A. M. Khan, S. H. Kee, A. S. K. Pathan and A. Al Nahid, Image processing techniques for concrete crack detection: A scientometrics literature review, Remote Sensing, 15(9), 2023, 1-46.
[17] A. Reghukumar and L. J. Anbarasi, Crack detection in concrete structures using deep learning, Lecture Notes in Electrical Engineering, 711, 2021, 211-219.
[18] M. R. Bandi, L. N. Pasupuleti, T. Das and S. Guchhait, Deep learning based damage detection of concrete structures, Asian Journal of Civil Engineering, 25, 2024,5197-5204.
[19] M. Mazni, A. R. Husain, M. I. Shapiai, I. S. Ibrahim, D. W. Anggara and R. Zulkifli, An investigation into real-time surface crack classification and measurement for structural health monitoring using transfer learning convolutional neural networks and Otsu method, Alexandria Engineering Journal, 92, 2024, 310-320.
[20] M. M. Islam, M. B. Hossain, M. N. Akhtar, M. A. Moni and K. F. Hasan, CNN based on transfer learning models using data augmentation and transformation for detection of concrete crack, Algorithms, 15(8), 2022, 287.
[21] A. P. Wibowo, A. Adha, I. F. Kurniawan and I. Laory, Wall crack multiclass classification: expertise-based dataset construction and learning algorithms performance comparison, Buildings, 12(12), 2022, 2135.
[22] D. W. Anggara et al., Concrete crack detection, orientation and measurement using a wall climbing robot, Applications of Modelling and Simulation, 8(1), 2024, 272-282.
[23] M. Mazni, A. R. Husain, M. I. Shapiai, I. S. Ibrahim, R. Zulkifli and D. W. Anggara, Real-time crack classification with wall-climbing robot using MobileNetV2, Communications in Computer and Information Series: Methods and Applications for Modeling and Simulation of Complex Systems, 1911, 2024, 319-328.
[24] R. Zulkifli, A. R. Husain, I. S. Ibrahim, M. Mazni and N. H. A. M. Fauzan, Analysis of the hybrid adhesion mechanism of the wall climbing robot, Lecture Notes in Electrical Engineering: Control, Instrumentation and Mechatronics: Theory and Practice, 921, 2022, 155-169.
[25] F. Sukarman, M. A. Mat Shah, M. Mazni, N. Abdul Malik and M. G. Mohd Hamami, Integrated concurrent programmable logic controller (PLC) based IR 4.0 training kit, The 9th International Innovation, Invention and Design Competition 2020, 2020, 360-364.
[26] F. Özgenel and A. Gönenç Sorguç, Performance comparison of pretrained convolutional neural networks on crack detection in buildings, ISARC 2018 - 35th International Symposium on Automation and Robotics in Construction and International AEC/FM Hackathon: The Future of Building Things, 2018.
[27] E. Bianchi and M. Hebdon, Development of Extendable Open-Source Structural Inspection Datasets, 36, 2022.
[28] Y. Liu, J. Yao, X. Lu, R. Xie and L. Li, DeepCrack: A deep hierarchical feature learning architecture for crack segmentation, Neurocomputing, 338, 2019, 139-153.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 Mazleenda Mazni, Abdul Rashid Husain, Devi Willieam Anggara, Mohd Ibrahim Shapiai, Riyadh Zulkifli, Munirah Onn
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:Authors hold and retain copyright, and grant the journal right of first publication, with the work after publication simultaneously licensed under a Creative Commons Attribution 4.0 License CC BY that permits any use, reproduction and distribution of the work and article without further permission provided that the original work is properly cited.
Authors are permitted and encouraged to post their work online in institutional repositories, website and other social media before and after publication, as it can lead to productive exchanges, as well as earlier and greater citation of published work.
