Detection of existing reinforced concrete structural irregularities using artificial intelligence and image processing techniques

In this study, structural irregularities such as B2 soft story irregularity and A3 plan protrusion were detected and classified using image processing and artificial intelligence methods. These irregularities in structures are of great importance, especially in terms of earthquake resistance and structural safety. Technological advancements allow for faster and more precise analyses compared to traditional methods in building inspections. Field observations and data collection were carried out using unmanned aerial vehicles (UAVs). This method provides an opportunity to examine structures with irregularities in detail through high-resolution images. The obtained images were then processed and analyzed using image processing techniques. Following the image processing steps, different artificial intelligence models were applied to determine the most suitable model for detecting irregularities. At the end of the study, the model that provided the most accurate results was selected and used for irregularity classification. With this approach, the aim is to bring a more systematic and technological dimension to structural safety analyses.