Ancient Building Crack Detection Based on YOLOv8 Algorithm

Main Article Content

Wei Xiong, Wei Meng, Zhanhao Liang, Chu Qiu, Olga Volodymyrivna Volichenko, Filatova Tatyana Arkadyevna

Abstract

This study investigates the use of the YOLOv8 algorithm for detecting cracks in ancient buildings. Utilizing a dataset from Kaggle, the model was trained to identify crack patterns with high accuracy. The YOLOv8 model achieved a precision of 1.00 at a confidence level of 0.823 and an overall accuracy of approximately 92%. These results demonstrate the effectiveness of YOLOv8 in accurately detecting and monitoring structural cracks, making it a valuable tool for the preservation of culturally significant structures. Future research will aim to enhance the model's capabilities and explore its integration with automated inspection technologies.

Article Details

Section
Articles