EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Lightweight One-Class Industrial Anomaly Detection Using a Normal-Calibrated Random-Subspace Ensemble

Çağatay Ersin

Electronics · 17.08.2026 · 15 (16) · 3654

DOI: 10.3390/electronics15163654 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Electronics, online erschienen am 17.08.2026. Verfasst von Çağatay Ersin. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Industrial anomaly detection is commonly performed under one-class conditions because defective samples are scarce or unavailable during training. This study presents a lightweight one-class anomaly detection framework that reduces the dependence of compact deep representations on a single arbitrarily selected feature subset. Intermediate features extracted from a frozen MobileNetV2 backbone are divided into five independently sampled 100-channel subspaces. A position-specific diagonal Gaussian model is fitted separately to each subspace using only normal training images. The resulting anomaly scores are robustly normalized and combined through a normal-calibrated reliability-weighting strategy, while image-level decisions are obtained from the highest-scoring local patches. To prevent information leakage, operating thresholds are determined through five-fold cross-fitted calibration using only normal training scores. The proposed compact random-subspace ensemble was evaluated on all 15 categories of MVTec AD and externally validated on the 12-category VisA benchmark. On MVTec AD, the ensemble increased the macro ROC-AUC from 0.9350 to 0.9437 and the macro F1 score from 0.8366 to 0.8504 compared with a single random 100-channel subspace. External validation on VisA showed the same improvement trend, with higher macro ROC-AUC, AP, and F1 values than the single-subspace configuration. The results demonstrate that combining multiple compact normal-data subspaces improves discrimination and reduces reliance on the evaluated single random representation while retaining a lightweight, anomaly-label-free training procedure.

Zitieren

APA

Ersin, Ç. (2026). Lightweight One-Class Industrial Anomaly Detection Using a Normal-Calibrated Random-Subspace Ensemble. Electronics, 15(16), 3654. https://doi.org/10.3390/electronics15163654

Vancouver

Ersin Ç. Lightweight One-Class Industrial Anomaly Detection Using a Normal-Calibrated Random-Subspace Ensemble. Electronics. 2026;15(16):3654. doi:10.3390/electronics15163654

BibTeX
@article{Ersin2026,
  author = {Ersin, Çağatay},
  title = {Lightweight One-Class Industrial Anomaly Detection Using a Normal-Calibrated Random-Subspace Ensemble},
  journal = {Electronics},
  year = {2026},
  volume = {15},
  number = {16},
  pages = {3654},
  doi = {10.3390/electronics15163654}
}
Kostenlos starten →