A Two-Stage Deep Learning Framework for Automated Corrosion Detection and Severity Estimation in High-Resolution SEM Images
Automation · 20.04.2026 · 7 (2) · 65
DOI: 10.3390/automation7020065 ↗Kurz gefasst
Fachartikel in der Zeitschrift Automation, online erschienen am 20.04.2026. Verfasst von Satyabrata Aich und 10 weiteren Autor:innen. Die Arbeit nennt 33 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Ministry of Education.
Abstract (Originalsprache)
Accurate detection and severity estimation of corrosion on metallic surfaces is essential for maintaining material integrity and ensuring operational safety in industrial systems. To address limitations in manual inspection methods, this study presents a two-stage deep learning pipeline tailored for high-resolution scanning electron microscopy images. The framework combines instance-level corrosion segmentation using the YOLOv8-seg architecture with subsequent severity classification performed by EfficientNet-B0 and ResNet18. In the segmentation stage, models are trained using both manually annotated and automatically generated binary masks, enabling robust instance mask prediction through prototype-based mask decoding. The classification stage assesses the severity of corrosion by analyzing localized regions based on morphological features, leveraging convolutional neural networks optimized for binary output. The experimental results demonstrate strong performance: the segmentation model trained on manual annotations achieves a Mean Intersection over Union (mIoU) of 89.91, a mask mAP@50 of 98.6, and an ROC-AUC of 94.69. For severity classification, EfficientNet-B0 achieves an accuracy of 93.75% and an F1-score of 93.29, outperforming ResNet18. The proposed framework connects advanced SEM with state-of-the-art machine learning. It provides a scalable, annotation-efficient way to use intelligent and automated corrosion characterization in materials science and industrial applications.
Förderung
- Ministry of Education
Zitieren
APA
Aich, S., Mohapatra, S., Nanda, S., Khan, T., Bharti, A., Sultana, H., Kalaiarsan, U., Senghuy, C., Ada, O. U. E., Mondal, P. K., & Lee, Y. (2026). A Two-Stage Deep Learning Framework for Automated Corrosion Detection and Severity Estimation in High-Resolution SEM Images. Automation, 7(2), 65. https://doi.org/10.3390/automation7020065
Vancouver
Aich S, Mohapatra S, Nanda S, Khan T, Bharti A, Sultana H, et al. A Two-Stage Deep Learning Framework for Automated Corrosion Detection and Severity Estimation in High-Resolution SEM Images. Automation. 2026;7(2):65. doi:10.3390/automation7020065
BibTeX
@article{Aich2026,
author = {Aich, Satyabrata and Mohapatra, Sudipta and Nanda, Shrabani and Khan, Taqdees and Bharti, Ayushi and Sultana, Hajra and Kalaiarsan, Umashankari and Senghuy, Chea and Ada, Okpete Uchenna Esther and Mondal, Proloy Kumar and Lee, Yong-Ki},
title = {A Two-Stage Deep Learning Framework for Automated Corrosion Detection and Severity Estimation in High-Resolution SEM Images},
journal = {Automation},
year = {2026},
volume = {7},
number = {2},
pages = {65},
doi = {10.3390/automation7020065}
}
EUVIMED