EUVIMEDEUVIMED entdecken
Crossref

Continual learning for surface defect segmentation by subnetwork creation and selection

Aleksandr Dekhovich, Miguel A. Bessa

Journal of Intelligent Manufacturing · 2025-06 · 36 (5) · 3051-3065

DOI: 10.1007/s10845-024-02393-4 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Intelligent Manufacturing, online erschienen am 06.05.2024. Verfasst von Aleksandr Dekhovich und 1 weiteren Autor:innen. Die Arbeit wurde bisher 8-mal zitiert (Crossref) und nennt 72 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

We introduce a new continual (or lifelong) learning algorithm called LDA-CP &S that performs segmentation tasks without undergoing catastrophic forgetting. The method is applied to two different surface defect segmentation problems that are learned incrementally, i.e., providing data about one type of defect at a time, while still being capable of predicting every defect that was seen previously. Our method creates a defect-related subnetwork for each defect type via iterative pruning and trains a classifier based on linear discriminant analysis (LDA). At the inference stage, we first predict the defect type with LDA and then predict the surface defects using the selected subnetwork. We compare our method with other continual learning methods showing a significant improvement – mean Intersection over Union better by a factor of two when compared to existing methods on both datasets. Importantly, our approach shows comparable results with joint training when all the training data (all defects) are seen simultaneously.

Zitieren

APA

Dekhovich, A., & Bessa, M. A. (2025). Continual learning for surface defect segmentation by subnetwork creation and selection. Journal of Intelligent Manufacturing, 36(5), 3051-3065. https://doi.org/10.1007/s10845-024-02393-4

Vancouver

Dekhovich A, Bessa MA. Continual learning for surface defect segmentation by subnetwork creation and selection. Journal of Intelligent Manufacturing. 2025;36(5):3051-3065. doi:10.1007/s10845-024-02393-4

BibTeX
@article{Dekhovich2025,
  author = {Dekhovich, Aleksandr and Bessa, Miguel A.},
  title = {Continual learning for surface defect segmentation by subnetwork creation and selection},
  journal = {Journal of Intelligent Manufacturing},
  year = {2025},
  volume = {36},
  number = {5},
  pages = {3051--3065},
  doi = {10.1007/s10845-024-02393-4}
}
Kostenlos starten →