Search research articles

ABOUT JoVE

Overview Leadership Blog JoVE Help Center

AUTHORS

Publishing Process Editorial Board Scope & Policies Peer Review FAQ Submit

LIBRARIANS

Testimonials Subscriptions Access Resources Library Advisory Board FAQ

RESEARCH

JoVE Journal Methods Collections JoVE Encyclopedia of Experiments Archive

EDUCATION

JoVE Core JoVE Business JoVE Science Education JoVE Lab Manual Faculty Resource Center Faculty Site

Terms & Conditions of Use

Related Concept Videos

Corrosion

Corrosion

The degradation of metals due to natural electrochemical processes is known as corrosion. Rust formation on iron, tarnishing of silver, and the blue-green patina that develops on copper are examples of corrosion. Corrosion involves the oxidation of metals. Sometimes it is protective, such as the oxidation of copper or aluminum, wherein a protective layer of metal oxide or its derivatives forms on the surface, protecting the underlying metal from further oxidation. In other cases, corrosion is...

Corrosion of Reinforcement

Corrosion of Reinforcement

The corrosion of steel reinforcement within concrete is a process influenced by the material's inherent properties and external factors. The high pH level of around 13, provided by calcium hydroxide present in concrete, initially protects the steel reinforcement by promoting the formation of a passive iron oxide layer on its surface.
However, over time and under certain conditions like carbonation, chloride ingress, and cracking this protective state can be compromised. Steel has areas with...

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by

Same author

CXCR2 blockade potentiates HDAC3 deletion mediated antitumor immunity in colorectal cancer.

Cell death & disease·2026

Same author

Mechanism-based subtypes of problematic use of the internet and corresponding neurobehavioral characteristics among children and adolescents.

Journal of child psychology and psychiatry, and allied disciplines·2026

Same author

CCR5+ CD8+ T Cells Are Associated with Poor Response to PD-1 Blockade Therapy.

International journal of molecular sciences·2026

Same author

Research on Quenching of 65Mn Friction Plates in Internal-Circulation Water Channel Molds Based on Finite Element Simulation.

Materials (Basel, Switzerland)·2026

Same author

Lactate Uptake by MCT4 Facilitates Stability and Suppressive Function of Tumor-Infiltrating Regulatory T Cells by Promoting Foxp3 Lactylation.

International journal of molecular sciences·2026

Same author

Restorative campus landscapes and student engagement: Dual affective pathways linking environmental perception to learning and social interaction.

Acta psychologica·2026

Search research articles

Related Experiment Video

Updated: Jun 13, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

Corrosion image classification method based on EfficientNetV2.

Ziheng Zhao¹, Elmi Bin Abu Bakar¹, Norizham Bin Abdul Razak¹

¹School of Aerospace Engineering, Kampus Kejuruteraan, Universiti Sains Malaysia, 14300, Nibong Tebal, Pulau Pinang, Malaysia.

|September 17, 2024

Summary

This study introduces a CNN model using transfer learning to detect material corrosion, achieving high accuracy. The model effectively identifies corrosion in various materials, enhancing safety assessments for critical infrastructure.

Keywords:

Accuracy CNN Corrosion EfficientNetV2 Fine-tuning Image classification

More Related Videos

Imaging Corrosion at the Metal-Paint Interface Using Time-of-Flight Secondary Ion Mass Spectrometry

Imaging Corrosion at the Metal-Paint Interface Using Time-of-Flight Secondary Ion Mass Spectrometry

Published on: May 6, 2019

Deep Neural Networks for Image-Based Dietary Assessment

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

Related Experiment Videos

Last Updated: Jun 13, 2025

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Author Spotlight: Enhancement of Salient Object Detection for Smart Grid Applications

Published on: December 15, 2023

Imaging Corrosion at the Metal-Paint Interface Using Time-of-Flight Secondary Ion Mass Spectrometry

Imaging Corrosion at the Metal-Paint Interface Using Time-of-Flight Secondary Ion Mass Spectrometry

Published on: May 6, 2019

Deep Neural Networks for Image-Based Dietary Assessment

Deep Neural Networks for Image-Based Dietary Assessment

Published on: March 13, 2021

Area of Science:

Materials Science
Computer Science
Engineering

Background:

Corrosion is a significant cause of material failure in critical infrastructure, posing risks to public safety and property.
Effective identification of corrosion across diverse materials and facilities is crucial for proactive maintenance and safety.
Current methods may lack the precision and scalability needed for comprehensive corrosion monitoring.

Purpose of the Study:

To develop and evaluate a Convolutional Neural Network (CNN) classification model for accurate material corrosion detection.
To leverage transfer learning and fine-tuning techniques with the EfficientNetV2 architecture for enhanced performance.
To investigate the impact of different pooling layers and fine-tuning strategies on corrosion classification accuracy.

Main Methods:

Utilized a corrosion binary classification dataset with data augmentation to improve robustness.
Employed transfer learning by fine-tuning various sizes of the EfficientNetV2 model.
Evaluated model performance using metrics such as Confusion Matrix, ROC curve, Precision, Recall, and F1-score.
Compared the efficacy of Global Average Pooling versus Global Max Pooling layers.

Main Results:

The Global Average Pooling layer demonstrated superior performance compared to Global Max Pooling.
EfficientNetV2B0 achieved the highest accuracy (0.9176) with a 20% fine-tuning rate.
EfficientNetV2S achieved a high ROC-AUC value of 0.97 with a 15% fine-tuning rate.
Precision, Recall, and F1-Score values consistently exceeded 0.9 for the best-performing models.

Conclusions:

The developed CNN model effectively detects and classifies material corrosion with high accuracy.
Transfer learning with EfficientNetV2, particularly using Global Average Pooling, offers a promising approach for corrosion identification.
The findings provide a valuable reference for developing advanced corrosion classification systems using EfficientNetV2.