The increasing challenges associated with corrosion in various industries have prompted the exploration of advanced techniques for detection and prediction. This review paper comprehensively examines the application of machine learning in corrosion detection. Leveraging a diverse set of datasets, including X-ray computed tomography (XCT), NEA, CAMCD, and a specialized dataset for Cross-country Pipeline Inspection Data Analysis and Testing of Probabilistic Degradation Models, we explore the efficacy of machine learning algorithms. The review highlights segmentation and classification techniques, emphasizing their role in accurate corrosion identification. Insights derived from this review contribute to the ongoing efforts to enhance corrosion monitoring and preventive strategies.
Paper
Full text
Applications of Machine Learning in Corrosion Detection
Semantic Scholar · Materials Science · 2024
Abstract
The increasing challenges associated with corrosion in various industries have prompted the exploration of advanced techniques for detection and prediction. This review paper comprehensively examines the application of machine learning in corrosion detection. Leveraging a diverse set of datasets, including X-ray computed tomography (XCT), NEA, CAMCD, and a specialized dataset for Cross-country Pipeline Inspection Data Analysis and Testing of Probabilistic Degradation Models, we explore the efficacy of machine learning algorithms. The review highlights segmentation and classification techniques, emphasizing their role in accurate corrosion identification. Insights derived from this review contribute to the ongoing efforts to enhance corrosion monitoring and preventive strategies.