Downloads: 1
Research Paper | Civil Engineering | Volume 15 Issue 4, April 2026 | Pages: 1601 - 1606 | India
BridgeGuard-Lite: A Low-Cost Image-Based Framework for Surface Crack Detection in Steel Structures
Abstract: Surface cracks in steel structures are critical indicators of structural degradation caused by fatigue, corrosion, and stress concentration. Early detection of such defects is essential for ensuring the safety and reliability of infrastructure systems. This paper proposes BridgeGuard-Lite, a simplified and cost-effective framework for detecting surface cracks in steel structures using basic image processing techniques. The proposed method utilises grayscale conversion, noise reduction, thresholding, and edge detection to identify crack patterns from steel surface images. A parameter adjustment step, including contrast enhancement, shape filtering, and minimum defect size constraints, is incorporated to improve detection performance under varying surface conditions. The effectiveness of the method is evaluated using object-based comparison and tuning-based analysis. The results demonstrate that the system successfully detects primary crack features and that parameter tuning significantly improves detection quality by reducing noise and enhancing crack continuity. The proposed framework provides a practical and low-cost solution for structural inspection and is suitable for preliminary condition assessment. Future enhancements may include integration of machine learning techniques and larger datasets to improve robustness and adaptability across diverse real-world conditions.
Keywords: Steel Crack Detection, Image Processing, Structural Health Monitoring, Bridge Monitoring, Automated Defect Detection
How to Cite?: Manjunath Gunjikar, Seema Narain, Dhanashree Tulankar, "BridgeGuard-Lite: A Low-Cost Image-Based Framework for Surface Crack Detection in Steel Structures", Volume 15 Issue 4, April 2026, International Journal of Science and Research (IJSR), Pages: 1601-1606, https://www.ijsr.net/getabstract.php?paperid=SR26424231009, DOI: https://dx.dx.doi.org/10.21275/SR26424231009