Advancing Robust Few-shot Surface Defect Detection through Meta-learning

Wenny Ramadha Putri, Yung-Hui Li, Jia Ching Wang

研究成果: 書貢獻/報告類型會議論文篇章同行評審

摘要

The primary limitation of deep learning (DL) techniques in detecting manufacturing defects is the requirement for a substantial volume of labeled data. Nevertheless, obtaining precise annotations is an arduous process due to the diverse range of defect categories. To tackle this issue, we propose MetaFormer, a transformer-based few-shot surface defect detection that learns the meta-features from a large-scale dataset of base classes. This method enables generalization to a novel class with only a few training examples. We demonstrate the effectiveness of our model on the steel surface defect dataset, surpassing the performance of the baseline model despite using only a limited number of annotated examples.

原文???core.languages.en_GB???
主出版物標題Proceedings of the 2024 9th International Conference on Integrated Circuits, Design, and Verification, ICDV 2024
編輯Xuan-Tu Tran, Duy-Hieu Bui
發行者Institute of Electrical and Electronics Engineers Inc.
頁面45-48
頁數4
ISBN(電子)9798350371864
DOIs
出版狀態已出版 - 2024
事件9th International Conference on Integrated Circuits, Design, and Verification, ICDV 2024 - Hanoi, Viet Nam
持續時間: 6 6月 20248 6月 2024

出版系列

名字Proceedings of the 2024 9th International Conference on Integrated Circuits, Design, and Verification, ICDV 2024

???event.eventtypes.event.conference???

???event.eventtypes.event.conference???9th International Conference on Integrated Circuits, Design, and Verification, ICDV 2024
國家/地區Viet Nam
城市Hanoi
期間6/06/248/06/24

指紋

深入研究「Advancing Robust Few-shot Surface Defect Detection through Meta-learning」主題。共同形成了獨特的指紋。

引用此