BROWSE
Detail
Conference Paper
Detection of Structural Damages for Petroglyphs of Bangudae Terrace using Edge Extraction based on Deep Learning
- Authors
- Sang-Yun Lee, Dong-Jun Shin
- Issue Date
- 2025-11
- Citation
- International Conference on Advances in Artificial Intelligence (ICAAI) 2025, pp.173-178
- Publisher
- ACM
- Language
- English
- Type
- Conference Paper
- Abstract
- We will deal with a case of applying Deep Learning technology to the conservation and management of the Petroglyphs of Bangudae Terrace in Daegok-ri, Ulju that were designated as national treasures. They were also selected as UNESCO World Heritage priority candidates, but have been continuously under threat of damage due to various environmental factors. In this paper, we attempted displacement detection using three Deep Learning models for efficient protection of the petroglyphs. The experimental results showed that the average processing times of the DexiNed and PiDiNet models were 6.43 and 20.87 seconds, respectively, which were effective for fast displacement detection, and the DeepCrack model was 114.19 seconds, which was suitable for large-scale data set analysis. In addition, the displacement detection results of the DeepCrack model confirmed that the joint area significantly changed when the temperature changed rapidly.
- Keyword
- Petroglyphs of Bangudae Terrace, Deep Learning, Displacement Detection, DexiNed, PiDiNet, DeepCrack, Cultural Heritage
- KSP Keywords
- Cultural Heritage, Edge extraction, Environmental Factors, Large-scale datasets, Learning Technology, UNESCO world heritage, deep learning(DL), deep learning models, displacement detection, structural damage
This work is distributed under the term of Creative Commons License (CCL)
(CC BY)
(CC BY)
