PSI - Issue 70

Bagathi HarshaVardhan et al. / Procedia Structural Integrity 70 (2025) 447–452

452

ResNet50 stands out in corrosion detection due to its deep residual learning framework, which effectively addresses the vanishing gradient problem in deep networks. This allows it to extract more complex features and achieve higher accuracy compared to simpler models like VGGNet. The system improves the reliability of corrosion severity assessments, lowers manual inspection costs, and enhances safety in industrial settings. VGGNet16 model is exhibiting very low accuracy for the corrosion assessment with the given dataset. The accuracy of the model is dependent on various parameters and also on the quality and complexity of imagery data. 5. Future Work: Future work could focus on balancing the dataset with techniques like SMOTE and better augmentation to improve accuracy. Using transfer learning, ensemble models, and domain adaptation can also help make the model more reliable and adaptable. References: Alapati, M., Ramesh Chandra, G. & Abboju, S.(2025. Efficacy of image-based deep learning CNN models for bolt loosening detection in steel structures. Asian Journal of Civil Engineering. https://doi.org/10.1007/s42107-025-01316-9 Ameli, Z., Jafarpoor Nesheli, S., & Landis, E. N. 2024. Deep Learning-Based Steel Bridge Corrosion Segmentation and Condition Rating Using Mask RCNN and YOLOv8. Infrastructures, 9(1), 3. https://doi.org/10.3390/infrastructures9010003 Bastian, N., et al. 2019. Visual inspection and characterization of external corrosion using deep learning. NDT & E International, 107, 102145. https://doi.org/10.1016/j.ndteint.2019.102134 Bolton, A., et al. 2022. A comparison of deep learning techniques for corrosion detection. Journal of Marine Science and Engineering, 10(11), 1625. https://doi.org/10.1007/978-3-031-20601-6_18 Cha, Y.-J., et al. 2018. Autonomous structural visual inspection using region-based deep learning for detecting multiple damage types. Computer Aided Civil and Infrastructure Engineering, 33(9), 731-747. https://doi.org/10.1111/mice.12334 Das, A., Dorafshan, S., & Kaabouch, N. 2024. Autonomous Image-Based Corrosion Detection in Steel Structures Using Deep Learning. Sensors (Basel, Switzerland), 24(11), 3630. https://doi.org/10.3390/s24113630 He, K., Zhang, X., Ren, S., & Sun, J. 2016. Deep Residual Learning for Image Recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 770-778. https://doi.org/10.1109/CVPR.2016.90 Holm, E., et al. 2019. Classification of corrosion and coating damages using CNNs. Journal of Structural Integrity and Maintenance, 4(4), 233-240. http://dx.doi.org/10.1117/12.2557380 Ma, Y. C., et al. 2018. Image-based corrosion recognition in ship steel structures. Journal of Marine Science and Technology, 23, 620-630. http://dx.doi.org/10.1117/12.2296540 Malik, Z., Mirani, A., Gopi, T. et al. 2024. A review on vision-based deep learning techniques for damage detection in bolted joints. Asian Journal of Civil Engineering, 25, 5697 – 5707. https://doi.org/10.1007/s42107-024-01139-0 Nabizadeh, A. H., & Parghi, A. 2023. Automated corrosion detection using deep learning and computer vision. Asian Journal of Civil Engineering, 24, 1-12. https://doi.org/10.1007/s42107-023-00684-4 Nash, W. T., et al. 2022. Deep learning corrosion detection with confidence. npj Materials Degradation, 6, 1-9. https://doi.org/10.1038/s41529 022-00232-6 Ta, Q. B., et al. 2022. Corroded bolt identification using deep learning. Automation in Construction, 135, 104126. https://doi.org/10.3390/22093340 Yao, Y., et al. 2019. CNN for corrosion detection in ship steel structures. Ocean Engineering, 186, 106131. https://doi.org/10.1016/j.apor.2019.05.008