Akademik Veri Yönetim Sistemi
4th International Civil Engineering & Architecture Conference, Trabzon, Türkiye, 17 - 19 Mayıs 2025, cilt.1, ss.218-226, (Tam Metin Bildiri)
This study aims to detect and evaluate damage in masonry structures using a contactless structural health monitoring method integrated with an Autoencoder deep learning model. Video recordings of a scaled minaret model constructed in a laboratory environment were taken at specific intervals using an industrial camera-based monitoring system. Subsequently, controlled damage was incrementally applied to the minaret model, and new video recordings were captured. Time-dependent displacement data from four different elevations of the scaled minaret model were extracted using a custom Python® script. A separate code based on an Autoencoder deep learning model was employed to calculate the damage index of the minaret model using the displacement data. Analyses conducted in the undamaged state revealed a stable damage index below 5%. However, analyses performed after the controlled damage steps showed that the damage index exceeded 5%. This study introduces a practical and rapid method for damage detection in masonry and similar engineering structures, facilitating faster diagnostic and decision-making processes in the field of structural health monitoring.