Comparative Analysis of VGG16 Transfer Learning Fine-Tuning Strategies for Automated Concrete Crack Classification
- Adwinof Akmal Juantoro Universitas Dian Nuswantoro, Semarang, Indonesia
-
Sugiyanto Sugiyanto
*
Universitas Dian Nuswantoro, Semarang,
Indonesia
(*) Corresponding Author
Abstract
Identifying cracks in concrete structures is critical for structural health monitoring, as undetected cracks can lead to catastrophic infrastructure failure. Conventional manual inspections are labour-intensive, subjective, and costly, necessitating automated solutions capable of consistent and scalable deployment. This paper presents a systematic comparative study of four VGG16 transfer learning strategies for automated binary classification of concrete surface cracks. VGG16 was selected for its proven effectiveness in binary image classification tasks, well-established pre-trained feature representations from ImageNet, and low trainable parameter count that reduces overfitting risk on domain-specific datasets. A dataset of 40,000 concrete surface photographs was utilised, divided 80:20 for training and validation. Four training configurations were evaluated: Baseline CNN, Full Freeze, Partial Fine-Tuning, and Full Fine-Tuning, all trained using the Adam optimiser (learning rate 0.001), binary cross-entropy loss, and early stopping. Partial Fine-Tuning achieved the highest accuracy at 99.90%, followed by Full Freeze (99.84%) and Baseline CNN (99.69%). Full Fine-Tuning collapsed to 50.00% due to catastrophic forgetting. The best-performing Partial Fine-Tuning configuration achieved an AUC of 0.9998, precision of 0.9990, recall of 0.9990, and F1-score of 0.9990, with only 15 misclassifications out of 8,000 validation samples. These results confirm that Partial Fine-Tuning is the recommended strategy for concrete crack classification in structural health monitoring application.
Downloads
References
P. Kumar, S. Batchu, N. S. S, and S. R. Kota, “Real-Time Concrete Damage Detection Using Deep Learning for High Rise Structures,” IEEE Access, vol. 9, pp. 112312–112331, 2021, doi: 10.1109/ACCESS.2021.3102647.
H. S. Munawar, F. Ullah, D. Shahzad, A. Heravi, S. Qayyum, and J. Akram, “Civil Infrastructure Damage and Corrosion Detection: An Application of Machine Learning,” Buildings, vol. 12, no. 2, 2022, doi: 10.3390/buildings12020156.
M. Panta, M. T. Hoque, M. Abdelguerfi, and M. C. Flanagin, “IterLUNet: Deep Learning Architecture for Pixel-Wise Crack Detection in Levee Systems,” IEEE Access, vol. 11, no. January, pp. 12249–12262, 2023, doi: 10.1109/ACCESS.2023.3241877.
Z. X. Zhang, H. L. Zhang, and T. J. Zhang, “Enhanced YOLOv8-based pavement crack detection: A high-precision approach,” PLoS One, vol. 20, no. 5 MAY, pp. 1–18, 2025, doi: 10.1371/journal.pone.0324512.
S. Katsigiannis, S. Seyedzadeh, A. Agapiou, and N. Ramzan, “Deep learning for crack detection on masonry façades using limited data and transfer learning,” J. Build. Eng., vol. 76, no. March, p. 107105, 2023, doi: 10.1016/j.jobe.2023.107105.
S. Roy, B. Yogi, R. Majumdar, P. Ghosh, and S. K. Das, “Deep learning-based crack detection and prediction for structural health monitoring,” Discov. Appl. Sci., vol. 7, no. 7, 2025, doi: 10.1007/s42452-025-07272-y.
A. M. Mayya and N. F. Alkayem, “Enhance the Concrete Crack Classification Based on a Novel Multi-Stage YOLOV10-ViT Framework,” Sensors, vol. 24, no. 24, 2024, doi: 10.3390/s24248095.
Y. Choi, “Application of Mask R-CNN and YOLOv8 Algorithms for Concrete Crack Detection,” IEEE Access, vol. 12, no. November, pp. 165314–165321, 2024, doi: 10.1109/ACCESS.2024.3469951.
A. Ashraf, A. Sophian, A. A. Shafie, T. S. Gunawan, N. N. Ismail, and A. A. Bawono, “Efficient Pavement Crack Detection and Classification Using Custom YOLOv7 Model,” Indones. J. Electr. Eng. Informatics, vol. 11, no. 1, pp. 119–132, 2023, doi: 10.52549/ijeei.v11i1.4362.
K. Zeng, R. Fan, and X. Tang, “Efficient and accurate road crack detection technology based on YOLOv8-ES,” Auton. Intell. Syst., vol. 3, 2025, doi: 10.1007/s43684-025-00091-3.
X. Wang, F. Zhang, and X. Zou, “Efficient Lightweight CNN and 2D Visualization for Concrete Crack Detection in Bridges,” Buildings, vol. 15, no. 18, p. 3423, Sep. 2025, doi: 10.3390/buildings15183423.
V. P. Golding, Z. Gharineiat, H. S. Munawar, and F. Ullah, “Crack Detection in Concrete Structures Using Deep Learning,” Sustainability, vol. 14, no. 13, p. 8117, Jul. 2022, doi: 10.3390/su14138117.
T. Fatima and H. Soliman, “Application of VGG16 Transfer Learning for Breast Cancer Detection,” Inf., vol. 16, no. 3, 2025, doi: 10.3390/info16030227.
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, vol. 15, no. 8, p. 287, Aug. 2022, doi: 10.3390/a15080287.
M. Shahin, F. F. Chen, M. Maghanaki, A. Hosseinzadeh, N. Zand, and H. Khodadadi Koodiani, “Improving the Concrete Crack Detection Process via a Hybrid Visual Transformer Algorithm,” Sensors, vol. 24, no. 10, p. 3247, May 2024, doi: 10.3390/s24103247.
W. Bukaita, K. Vankudothu, and J. Khan, “Automated Multi-Class Concrete Crack Detection and Severity Classification Using CNN-Based Deep Learning,” Am. J. Civ. Eng., vol. 13, no. 4, pp. 197–210, Jul. 2025, doi: 10.11648/j.ajce.20251304.12.
L. Ali, F. Alnajjar, H. Al Jassmi, M. Gocho, W. Khan, and M. A. Serhani, “Performance Evaluation of Deep CNN-Based Crack Detection and Localization Techniques for Concrete Structures,” 2021.
P. Jing, H. Yu, Z. Hua, S. Xie, and C. Song, “Road Crack Detection Using Deep Neural Network Based on Attention Mechanism and Residual Structure,” IEEE Access, vol. 11, pp. 919–929, 2023, doi: 10.1109/ACCESS.2022.3233072.
E. Wahyudi, B. Imran, A. Subki, Z. Zaeniah, L. D. Samsumar, and S. Salman, “Crack Detection of Concrete Surfaces with A Combination of Feature Extraction and Image-Based Backpropagation Artificial Neural Networks,” Ilk. J. Ilm., vol. 16, no. 3, pp. 228–235, 2024, doi: 10.33096/ilkom.v16i3.2249.228-235.
A. Altaf, A. Mehmood, M. L. Filograno, S. Alharbi, and J. Iqbal, “Deployable Deep Learning Models for Crack Detection: Efficiency, Interpretability, and Severity Estimation,” Buildings, vol. 15, no. 18, pp. 1–21, 2025, doi: 10.3390/buildings15183362.
Bila bermanfaat silahkan share artikel ini
Berikan Komentar Anda terhadap artikel Comparative Analysis of VGG16 Transfer Learning Fine-Tuning Strategies for Automated Concrete Crack Classification
Pages: 2597-2608
Copyright (c) 2026 Adwinof Akmal Juantoro, Sugiyanto Sugiyanto
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 retain copyright and grant the journal right of first publication with the work simultaneously licensed under Creative Commons Attribution 4.0 International License that allows others to share the work with an acknowledgment of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (Refer to The Effect of Open Access).
