International Display Workshops General Incorporated Association

3:30 PM - 3:50 PM

[DES1-3L] Crack detection and Wooden Artifacts Preservation System Based on Random Forest Classifier and WGAN-GP Network

*YA CHUN CHOU1, Chung Wen Hung1 (1. National Yunlin University of Science and Technology (Taiwan))

defect detection, antiquities preservation, machine learning, WGAN-GP, random forest

https://doi.org/10.36463/idw.2023.1041

Defect detection with Machine learning algorithm is adopted to identify cracks on antiquities in this paper. Additionally, the WGAN-GP network is also adopted to generate defect samples for improving the results of machine learning. The experimental results show that the system is particularly effective, which can reach 0.95 for accuracy.