EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Addressing Data Scarcity in Additive Manufacturing Monitoring via Synthetic Data Generation and Meta Pseudo-Labeling for Foundational Layer-Wise Segmentation

Yie Sheng Chen, Petro Mushidi Tshakwanda, Henok Berhanu Tsegaye, Jin Zhang, Harsh Kumar, Michael Devetsikiotis

Journal of Manufacturing and Materials Processing · 27.05.2026 · 10 (6) · 183

DOI: 10.3390/jmmp10060183 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Manufacturing and Materials Processing, online erschienen am 27.05.2026. Verfasst von Yie Sheng Chen und 5 weiteren Autor:innen. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Science Foundation.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Additive manufacturing (AM) monitoring is fundamentally constrained by the severe scarcity of annotated data for layer-wise segmentation. This paper addresses this bottleneck by introducing a scalable, high-fidelity synthetic data generation pipeline built on the Slice-100K dataset, capable of producing large volumes of layer-wise semantic segmentation masks. Through analysis of this large-scale synthetic data, we identify a systemic foreground–background class imbalance (1:24 ratio) inherent to AM monitoring, which causes standard Dice loss formulations to diverge catastrophically into a phenomenon we formalize as the “Dice Crash.” To effectively leverage large amounts of unlabeled data, we adapt the Meta Pseudo-Labeling (MPL) framework for industrial segmentation. We evaluate MPL’s true marginal utility by integrating it with both a standard U-Net and a robust state-of-the-art nnU-Net architecture. Experimental outputs show that while MPL yields substantial performance gains (+15.2%) on weak baselines, integrating it with an optimally configured strong baseline consistently improves segmentation accuracy and suppresses false foreground detections, thereby mitigating confirmation bias. These findings demonstrate that semi-supervised learning via continuous bilevel optimization offers a practical and robust enhancement to data-scarce additive manufacturing monitoring. Because any hidden defects in the topmost layer will be permanently buried by subsequent extrusion, this foundational layer-wise segmentation step is the most critical primitive of the monitoring pipeline.

Förderung

  • National Science Foundation

Zitieren

APA

Chen, Y. S., Tshakwanda, P. M., Tsegaye, H. B., Zhang, J., Kumar, H., & Devetsikiotis, M. (2026). Addressing Data Scarcity in Additive Manufacturing Monitoring via Synthetic Data Generation and Meta Pseudo-Labeling for Foundational Layer-Wise Segmentation. Journal of Manufacturing and Materials Processing, 10(6), 183. https://doi.org/10.3390/jmmp10060183

Vancouver

Chen YS, Tshakwanda PM, Tsegaye HB, Zhang J, Kumar H, Devetsikiotis M. Addressing Data Scarcity in Additive Manufacturing Monitoring via Synthetic Data Generation and Meta Pseudo-Labeling for Foundational Layer-Wise Segmentation. Journal of Manufacturing and Materials Processing. 2026;10(6):183. doi:10.3390/jmmp10060183

BibTeX
@article{Chen2026,
  author = {Chen, Yie Sheng and Tshakwanda, Petro Mushidi and Tsegaye, Henok Berhanu and Zhang, Jin and Kumar, Harsh and Devetsikiotis, Michael},
  title = {Addressing Data Scarcity in Additive Manufacturing Monitoring via Synthetic Data Generation and Meta Pseudo-Labeling for Foundational Layer-Wise Segmentation},
  journal = {Journal of Manufacturing and Materials Processing},
  year = {2026},
  volume = {10},
  number = {6},
  pages = {183},
  doi = {10.3390/jmmp10060183}
}
Kostenlos starten →