Cracks Detection of Ancient Objects using Neural Network

Abstract

Preserving and maintaining the structural integrity of ancient architectural features is of crucial importance for cultural heritage conservation. Over time, these historical sites often acquire fissures and structural problems, offering considerable obstacles for restoration and protection initiatives. In this paper, we offer a novel strategy to solve the essential issue of crack detection in historic sites by employing Deep Convolutional Neural Networks (CNNs). To evaluate the model's performance, we conducted trials on a wide range of photos documenting ancient places from throughout the world. The application of deep CNNs in fracture detection for ancient places promises to be a valuable tool for cultural preservation, enabling more efficient and preventive maintenance measures.