The use of artificial intelligence in damage assessment of historical buildings

Abstract

t Historical buildings are indispensable for maintaining cultural continuity. Proper restoration is the only way to preserve their original character. At this stage, early and accurate diagnosis of the damage to the historical building plays a vital role in the restoration process. Traditional damage assessment methods sometimes cause erroneous diagnoses and damage to the building. For this reason, non-destructive methods should be developed by utilizing the opportunities provided by technology. The research aims to develop an artificial intelligence-based damage detection model that can quickly and accurately detect deterioration in historical buildings. The study’s scope consists of traditional Gaziantep houses in the city’s historical center. The primary materials are high-resolution digital fac¸ade images, survey reports of these houses, and the findings obtained in the field research. The research reveals that deterioration maps, which are prepared with traditional methods by spending intensive labor and time, can be produced with an artificial intelligence-based system. Experts first documented the damages seen on the fac¸ades of historic stone buildings, and the model trained with these data was used as a supportive method to determine the types of deterioration. Integrating the system with expert opinions, field studies, and visual documents makes creating deterioration maps more efficient. ©2025 The Author(s). Publishing services by Elsevier B.V. on behalf of Higher Education Press and KeAi. This is an open access article under the CC BY-NC-ND license.