EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Source-Model-Guided Target-Group Sampling for Label-Efficient Four-Level Building-Damage Adaptation in Conflict Regions

Na Li, Meng Ren, Gaijun Zhao, Lilu Zhu, Zhe Zhang

Applied Sciences · 17.11.2026 · 16 (22) · 11367

DOI: 10.3390/app162211367 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 17.11.2026. Verfasst von Na Li und 4 weiteren Autor:innen. Die Arbeit nennt 45 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Frontier Technologies R&D Program of Jiangsu.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Four-level building-damage assessment from paired satellite imagery requires extensive expert labeling, yet models trained on natural-disaster data transfer poorly to conflict imagery. We propose a source-model-guided framework coupling cross-hazard transfer learning with pre-annotation selection of target-domain image groups. An xBD-trained bi-temporal source classifier initializes conflict-domain fine-tuning and provides building-level predictions for group screening. Predictions are aggregated into a group-level Damage Severity Index (DSI) and predicted-label purity; DSI stratification allocates the budget across predicted severity levels, while purity differentiates groups within strata. In a retrospective label-hiding experiment on a Ukraine dataset, selecting 40 of 120 groups (525 of 1615 labels) reduced labeling by 67.49%. The model achieved 84.29% within-one-grade accuracy, 53.43% exact accuracy, 49.37% Macro-F1, and 0.662 ordinal MAE, close to using all groups. Compared with two uncertainty baselines, it yielded better Macro-F1 and ordinal error estimates at comparable exact accuracy. Together with evaluation in Mariupol, these results suggest that source-model-guided selection can reduce target-domain labeling needs while maintaining competitive performance. However, group selection depends on source-prediction quality, and validation remains limited to regions within one conflict dataset. Future work will investigate more robust screening descriptors and additional conflict datasets.

Förderung

  • Frontier Technologies R&D Program of Jiangsu

Zitieren

APA

Li, N., Ren, M., Zhao, G., Zhu, L., & Zhang, Z. (2026). Source-Model-Guided Target-Group Sampling for Label-Efficient Four-Level Building-Damage Adaptation in Conflict Regions. Applied Sciences, 16(22), 11367. https://doi.org/10.3390/app162211367

Vancouver

Li N, Ren M, Zhao G, Zhu L, Zhang Z. Source-Model-Guided Target-Group Sampling for Label-Efficient Four-Level Building-Damage Adaptation in Conflict Regions. Applied Sciences. 2026;16(22):11367. doi:10.3390/app162211367

BibTeX
@article{Li2026,
  author = {Li, Na and Ren, Meng and Zhao, Gaijun and Zhu, Lilu and Zhang, Zhe},
  title = {Source-Model-Guided Target-Group Sampling for Label-Efficient Four-Level Building-Damage Adaptation in Conflict Regions},
  journal = {Applied Sciences},
  year = {2026},
  volume = {16},
  number = {22},
  pages = {11367},
  doi = {10.3390/app162211367}
}
Kostenlos starten →