BROWSE
Detail
Conference Paper
인공지능 알고리즘을 이용한 흥인지문 지붕부 변위 판정 모델 성능 향상에 대한 연구
Cited - time in 
Share 
Share
- Issue Date
- 2023-06
- Citation
- 대한전자공학회 학술 대회 (하계) 2023, pp.1050-1054
- Publisher
- 대한전자공학회
- Language
- Korean
- Type
- Conference Paper
- Abstract
- Cultural heritage is exposed to the outside for a long time, so structural deformation, deterioration, and cracks easily occur, so it is necessary to investigate damage to the structure through regular safety inspections so that the original form of the cultural heritage can be well preserved. is limited, and there is a limit to quickly responding to it by identifying it. In this paper, an experiment was conducted to predict the slope of the roof of Heunginjimun Gate in various environments using 10 different pre-trained neural network models based on deep learning, and through parameter adjustment of the model whose prediction accuracy was less than 95%, We were able to increase the prediction accuracy to more than 95%, and present the analysis results.
- KSP Keywords
- Cultural Heritage, Long Time, Parameter Adjustment, Prediction accuracy, deep learning(DL), neural network model, structural deformations