107‐4: Open‐Set PCB Defect Detection for Dynamic Display Manufacturing
SID Symposium Digest of Technical Papers · 2026-06 · 57 (1) · 1495-1498
DOI: 10.1002/sdtp.70001_404 ↗Kurz gefasst
Fachartikel in der Zeitschrift SID Symposium Digest of Technical Papers, online erschienen am 11.09.2026. Verfasst von Qiao Xu und 3 weiteren Autor:innen. Die Arbeit nennt 7 Literaturangaben.
Abstract (Originalsprache)
Acknowledging the limitation of traditional defect detection algorithms in identifying novel and unknown defects in PCBs, this paper proposes an open‐set detection and identification method. This approach enhances defect representation through image registration and differential analysis, performs class‐agnostic localization of all potential defects based on features such as shape and location, and quantifies classification uncertainty during inference to distinguish between known and unknown defects. Evaluated on a public dataset, the method reduces the miss rate for unknown defects by 49.5%.
Zitieren
APA
Xu, Q., Huai, L., Liu, Y., & Jiang, X. (2026). 107‐4: Open‐Set PCB Defect Detection for Dynamic Display Manufacturing. SID Symposium Digest of Technical Papers, 57(1), 1495-1498. https://doi.org/10.1002/sdtp.70001_404
Vancouver
Xu Q, Huai L, Liu Y, Jiang X. 107‐4: Open‐Set PCB Defect Detection for Dynamic Display Manufacturing. SID Symposium Digest of Technical Papers. 2026;57(1):1495-1498. doi:10.1002/sdtp.70001_404
BibTeX
@article{Xu2026,
author = {Xu, Qiao and Huai, Lian and Liu, Yuyu and Jiang, Xingqun},
title = {107‐4: Open‐Set PCB Defect Detection for Dynamic Display Manufacturing},
journal = {SID Symposium Digest of Technical Papers},
year = {2026},
volume = {57},
number = {1},
pages = {1495--1498},
doi = {10.1002/sdtp.70001_404}
}
EUVIMED