Recently, visual inspection methods using deep learning have been proposed. In this paper, we propose a classification method using two-step deep learning. The first step is to determine whether the painting is a defect or not, and the second step is to determine the kind of painting defect. By comparing the results using various deep learning models, we show that the classification accuracy is higher than that of conventional method.
J-STAGEがリニューアルされました! https://www.jstage.jst.go.jp/browse/-char/ja/