Non-Destructive Image Processing Methodology for Monitoring Corrosion in Concrete Reinforcement
Nishant Kumar1, Prof. Satya Prakash2
1Department of Civil Engineering, Sharda University, Greater Noida.
2Department of Civil Engineering, Sharda University, Greater Noida
Abstract - Corrosion of reinforcement bars (rebars) is one of the most critical factors affecting the durability and safety of reinforced concrete (RC) structures. Conventional inspection techniques, such as half-cell potential testing, though reliable, are time-consuming, invasive, and provide limited spatial resolution. This study presents a non-destructive, automated approach for corrosion assessment using digital image processing algorithms implemented in Python with OpenCV. High-resolution images of rebars were pre-processed through resizing, blurring, grayscale conversion, thresholding, and contour detection, followed by polygon-based masking and color analysis to accurately quantify corroded regions. Validation was carried out against half-cell potential measurements (ASTM C876) across 20 reinforcement samples. Results demonstrated strong agreement between the two methods: corrosion levels ranged from 0.46% (-194 mV, low probability) to 97.53% (-610 mV, high probability), with consistent correlation across low, moderate, and severe damage ranges. The image processing algorithm achieved 100% true positive detection with no false classifications, while providing the added advantage of spatial visualization of corrosion patterns. The findings confirm that image-based assessment can complement traditional electrochemical techniques, improving diagnostic accuracy and efficiency. This approach offers a scalable, cost-effective, and non-invasive tool for structural health monitoring, with significant potential for integration into predictive maintenance and life-cycle management of RC structures.
Key Words: Digital Image Processing; Corrosion Detection; Reinforced Concrete; Half-Cell Potential; Structural Health Monitoring; Non-Destructive Testing; Python Algorithms.
