EUVIMEDEUVIMED entdecken
Open Access · CC BY-NC 4.0

Transfer Learning Neural Networks for Nuclear Forensic Image Morphology Using Image Splitting Techniques

Niko A. Petrocelli, Lee C. Lambert, Brett J. Borghetti, Abigail A. Bickley

Journal of Chemometrics · 2026-03 · 40 (3)

DOI: 10.1002/cem.70112 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Chemometrics, online erschienen am 09.03.2026. Verfasst von Niko A. Petrocelli und 3 weiteren Autor:innen. Die Arbeit nennt 21 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0). Gefördert von: U.S. Department of Defense.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Manual morphological analysis of actinide particles from scanning electron microscope (SEM) imagery is a critical component of nuclear forensics but is prone to significant inter‐analyst variability. To address this challenge, this work develops and evaluates an automated classification method using deep learning. We introduce a methodology based on partitioning 1906 SEM images, representing 13 classes of uranium compounds, into smaller patches for analysis. Three convolutional neural network (CNN) architectures of increasing complexity were compared: a custom baseline CNN, a simple transfer learning model using ResNet50v1, and a complex model featuring hierarchical feature extraction and a spatial attention mechanism built upon ResNet50v2. The final image classification was determined using a patch‐based, class‐balanced voting system. The complex model achieved a superior balanced accuracy of 94% on the test set, significantly outperforming the simple transfer learning model (89%) and the baseline model (77%). These results demonstrate that a sophisticated transfer learning architecture can serve as a robust, objective tool to increase the accuracy and consistency of actinide particle classification, thereby enhancing nuclear forensic capabilities.

Förderung

  • U.S. Department of Defense

Zitieren

APA

Petrocelli, N. A., Lambert, L. C., Borghetti, B. J., & Bickley, A. A. (2026). Transfer Learning Neural Networks for Nuclear Forensic Image Morphology Using Image Splitting Techniques. Journal of Chemometrics, 40(3). https://doi.org/10.1002/cem.70112

Vancouver

Petrocelli NA, Lambert LC, Borghetti BJ, Bickley AA. Transfer Learning Neural Networks for Nuclear Forensic Image Morphology Using Image Splitting Techniques. Journal of Chemometrics. 2026;40(3). doi:10.1002/cem.70112

BibTeX
@article{Petrocelli2026,
  author = {Petrocelli, Niko A. and Lambert, Lee C. and Borghetti, Brett J. and Bickley, Abigail A.},
  title = {Transfer Learning Neural Networks for Nuclear Forensic Image Morphology Using Image Splitting Techniques},
  journal = {Journal of Chemometrics},
  year = {2026},
  volume = {40},
  number = {3},
  doi = {10.1002/cem.70112}
}
Kostenlos starten →