Abstract
The study employs Python-based image processing and laboratory experiments to detect and monitor cracks in structures. It tracks crack growth, severity, and angle over time, offering a comprehensive analysis of each crack that forms on the surface of the concrete by forming segments. The program's capabilities include precise crack detection by four directional matrices operation and minimal error rates of crack development. The algorithm can be worked in any condition due to supervised automation. The mathematical formulation for threshold performs better than OTSU in the case of concrete surfaces. The actual crack length is calculated with the help of the tortuosity index, formulated with respect to the mean width of the crack. Crack width and crack length are both taken as parameters for severity based on which its level is decided. The total five grades of severity level are defined. This approach provides a systematic and organized method for monitoring and analyzing changes in cracks, which is crucial for maintaining structural safety.