カテゴリ: 論文誌(論文単位)グループ名: 【D】産業応用部門（英文）発行日: 2022/07/01タイトル(英語): Enhancing Anomaly Detection Performance and Acceleration著者名: Ryo Saiku (Department of Intelligence Science and Engineering, Graduate School of Natural Science and Technology, Gifu University), Junya Sato (Faculty of Engineering, Gifu University), Takayoshi Yamada (Faculty of Engineering, Gifu University), Kazuaki I著者名(英語): Ryo Saiku (Department of Intelligence Science and Engineering, Graduate School of Natural Science and Technology, Gifu University), Junya Sato (Faculty of Engineering, Gifu University), Takayoshi Yamada (Faculty of Engineering, Gifu University), Kazuaki Ito (Faculty of Engineering, Gifu University)キーワード: anomaly detection，deep learning，visual inspection要約(英語): Automation of visual inspection is a critical aspect in industrial fields. Recently, research on anomaly detection using neural networks has been gaining increasing attention. In particular, approaches that use a pre-trained convolutional neural network have exhibited high performance. In this study, we focused on PatchCore, which is a high-performance model, and further improved it using two high-resolution images to accurately detect small anomalies. However, the extracted features (memory bank) consume a large amount of memory and storage; the memory bank is compressed by k-means clustering. Moreover, the inference time was reduced by an approximate nearest-neighbor search using an inverted index. Our method achieved an image-level AUROC of 0.994 on the MVTec anomaly detection dataset. In addition, a pixel-level AUROC of 0.984 was achieved, which is better than that of PatchCore. Furthermore, the compression time was reduced by more than 97% by clustering the memory bank using k-means while maintaining the performance.本誌: IEEJ Journal of Industry Applications Vol.11 No.4 （2022）本誌掲載ページ: 616-622 p原稿種別: 論文／英語電子版へのリンク: https://www.jstage.jst.go.jp/article/ieejjia/11/4/11_21013871/_article/-char/ja/
