EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Weakly Supervised Fine-Grained Aircraft Detection in Remote Sensing Based on Prior-Knowledge Prototype Learning

Peng Chen, Yang Qu, Ruye Li, Zicong Zhu

Remote Sensing · 17.07.2026 · 18 (14) · 2390

DOI: 10.3390/rs18142390 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 17.07.2026. Verfasst von Peng Chen und 3 weiteren Autor:innen. Die Arbeit nennt 56 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China, Jiangsu Provincial Department of Education.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

The weakly supervised aircraft fine-grained detection task remains challenging due to the lack of fine-grained category labels. In addition, the high inter-class similarity among aircraft categories and complex background interference further hinders discriminative feature learning. To address these challenges, we propose a weakly supervised fine-grained aircraft detection framework based on prior-knowledge prototype learning. Specifically, a Prior-knowledge-based Soft-label Generator (PSG) is designed to exploit geometric and semantic priors, enabling the propagation of fine-grained category annotations from a few annotated samples to large-scale unlabeled data through reliable soft-label generation. To alleviate category ambiguity caused by highly similar aircraft categories, a Text Prototype Alignment (TPA) strategy is introduced to align vision-language semantic knowledge with the remote sensing domain and construct discriminative semantic prototypes. Furthermore, a Feature Purification Module (FPM) is developed to suppress background noise through morphology-guided feature purification and enhance aircraft-specific representations. Experiments on the FAIR1M-Aircraft dataset demonstrate that the proposed method achieves 30.5% mAP with only 50 fine-grained annotated images, outperforming the baseline by 10.1 mAP points.

Förderung

  • National Natural Science Foundation of China
  • Jiangsu Provincial Department of Education

Zitieren

APA

Chen, P., Qu, Y., Li, R., & Zhu, Z. (2026). Weakly Supervised Fine-Grained Aircraft Detection in Remote Sensing Based on Prior-Knowledge Prototype Learning. Remote Sensing, 18(14), 2390. https://doi.org/10.3390/rs18142390

Vancouver

Chen P, Qu Y, Li R, Zhu Z. Weakly Supervised Fine-Grained Aircraft Detection in Remote Sensing Based on Prior-Knowledge Prototype Learning. Remote Sensing. 2026;18(14):2390. doi:10.3390/rs18142390

BibTeX
@article{Chen2026,
  author = {Chen, Peng and Qu, Yang and Li, Ruye and Zhu, Zicong},
  title = {Weakly Supervised Fine-Grained Aircraft Detection in Remote Sensing Based on Prior-Knowledge Prototype Learning},
  journal = {Remote Sensing},
  year = {2026},
  volume = {18},
  number = {14},
  pages = {2390},
  doi = {10.3390/rs18142390}
}
Kostenlos starten →