Next Article in Journal
Computer Numerical Control Machining Process Simulation in Brownfield Environments: Digital Twin, Artificial Intelligence Optimisation, and Implementation Roadmap
Previous Article in Journal
From Virtual Trajectory Generation to Real Execution and Validation in a MATLAB-ROS Hybrid Framework for a 6 DOF Industrial Robot
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Automation
Volume 7
Issue 2
10.3390/automation7020065
automation-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
first_page
settings
Order Article Reprints
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

A Two-Stage Deep Learning Framework for Automated Corrosion Detection and Severity Estimation in High-Resolution SEM Images

by
Satyabrata Aich
1,
Sudipta Mohapatra
2,
Shrabani Nanda
3,
Taqdees Khan
4,
Ayushi Bharti
4,
Hajra Sultana
5,
Umashankari Kalaiarsan
4,
Chea Senghuy
4,
Okpete Uchenna Esther Ada
4,
Proloy Kumar Mondal
4 and
Yong-Ki Lee
6,*
1
Wellmatix Limited, Daejeon 34141, Republic of Korea
2
Division of Advanced Materials Engineering, Jeonbuk National University, Jeonju 54896, Republic of Korea
3
Department of Business Intelligence, Ajou University, Suwon 16499, Republic of Korea
4
Department of Digital Anti-Aging Healthcare, Inje University, Gimhae 50834, Republic of Korea
5
Department of Computer Engineering, Inje University, Gimhae 50834, Republic of Korea
6
Department of Design, Dongseo University, Busan 47011, Republic of Korea
*
Author to whom correspondence should be addressed.
Automation 2026, 7(2), 65; https://doi.org/10.3390/automation7020065
Submission received: 16 February 2026 / Revised: 22 March 2026 / Accepted: 14 April 2026 / Published: 20 April 2026
Browse Figures
Versions Notes

Abstract

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.
Keywords: corrosion detection; instance segmentation; YOLOv8; EfficientNet; SEM imagery; computer vision; CNN classification; industrial inspection corrosion detection; instance segmentation; YOLOv8; EfficientNet; SEM imagery; computer vision; CNN classification; industrial inspection

Share and Cite

MDPI and ACS Style

Aich, S.; Mohapatra, S.; Nanda, S.; Khan, T.; Bharti, A.; Sultana, H.; Kalaiarsan, U.; Senghuy, C.; Ada, O.U.E.; Mondal, P.K.; et al. A Two-Stage Deep Learning Framework for Automated Corrosion Detection and Severity Estimation in High-Resolution SEM Images. Automation 2026, 7, 65. https://doi.org/10.3390/automation7020065

AMA Style

Aich S, Mohapatra S, Nanda S, Khan T, Bharti A, Sultana H, Kalaiarsan U, Senghuy C, Ada OUE, Mondal PK, 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. https://doi.org/10.3390/automation7020065

Chicago/Turabian Style

Aich, Satyabrata, Sudipta Mohapatra, Shrabani Nanda, Taqdees Khan, Ayushi Bharti, Hajra Sultana, Umashankari Kalaiarsan, Chea Senghuy, Okpete Uchenna Esther Ada, Proloy Kumar Mondal, and et al. 2026. "A Two-Stage Deep Learning Framework for Automated Corrosion Detection and Severity Estimation in High-Resolution SEM Images" Automation 7, no. 2: 65. https://doi.org/10.3390/automation7020065

APA Style

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.-K. (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

Article Metrics

Back to TopTop