ResUnet-GAN with Dynamic Memory for Mura Defect Detection

Chia Yu Lin, Pin Fan Lin, Wei Kuang Chung, Yu Hsien Lee

研究成果: 書貢獻/報告類型會議論文篇章同行評審

摘要

'Mura' is a phenomenon in which panels have uneven display defects, irregular shapes, and different sizes. It is impossible to produce perfect panels on production lines, so panel inspection is necessary to differentiate between 'light Mura' and 'serious Mura' manually. The performance of conventional defect detection models for Mura detection is worse since they only differentiate between 'normal' and 'abnormal' samples. To reduce human cost and increase the accuracy of Mura detection, we propose a 'ResUnet-GAN with Dynamic Memory Model an unsupervised anomaly detection method based on a Generative Adversarial Network (GAN) with a memory module to distinguish panel defects. In the dynamic memory, we designed a dynamic feature filtering (DFF) method to choose important features of images, enhancing the ability to recognize light Mura features of the ResUnet-GAN. The proposed model can achieve an Area Under Curve (AUC) of approximately 0.8 for accurate Mura detection. The mechanism of this paper is novel, and the result contributes to practical application.

原文???core.languages.en_GB???
主出版物標題Proceedings of the 2023 IEEE International Conference on Industry 4.0, Artificial Intelligence, and Communications Technology, IAICT 2023
發行者Institute of Electrical and Electronics Engineers Inc.
頁面350-353
頁數4
ISBN(電子)9798350313635
DOIs
出版狀態已出版 - 2023
事件2023 IEEE International Conference on Industry 4.0, Artificial Intelligence, and Communications Technology, IAICT 2023 - Hybrid, Bali, Indonesia
持續時間: 13 7月 202315 7月 2023

出版系列

名字Proceedings of the 2023 IEEE International Conference on Industry 4.0, Artificial Intelligence, and Communications Technology, IAICT 2023

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

???event.eventtypes.event.conference???2023 IEEE International Conference on Industry 4.0, Artificial Intelligence, and Communications Technology, IAICT 2023
國家/地區Indonesia
城市Hybrid, Bali
期間13/07/2315/07/23

指紋

深入研究「ResUnet-GAN with Dynamic Memory for Mura Defect Detection」主題。共同形成了獨特的指紋。

引用此