EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

RAPT-Net: Reliability-Aware Precision-Preserving Tolerance-Enhanced Network for Tiny Target Detection in Wide-Area Coverage Aerial Remote Sensing

Peida Zhou, Xiaojun Guo, Xiaoyong Sun, Bei Sun, Shaojing Su, Wei Jiang, Runze Guo, Zhaoyang Dang, Siyang Huang

Remote Sensing · 01.02.2026 · 18 (3) · 449

DOI: 10.3390/rs18030449 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Remote Sensing, online erschienen am 01.02.2026. Verfasst von Peida Zhou und 8 weiteren Autor:innen. Die Arbeit wurde bisher 3-mal zitiert (Crossref) und nennt 55 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Hunan Provincial Postgraduate Research Innovation Programme.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Multi-platform aerial remote sensing supports critical applications including wide-area surveillance, traffic monitoring, maritime security, and search and rescue. However, constrained by observation altitude and sensor resolution, targets inherently exhibit small-scale characteristics, making small object detection a fundamental bottleneck. Aerial remote sensing faces three unique challenges: (1) spatial heterogeneity of modality reliability due to scene diversity and illumination dynamics; (2) conflict between precise localization requirements and progressive spatial information degradation; (3) annotation ambiguity from imaging physics conflicting with IoU-based training. This paper proposes RAPT-Net with three core modules: MRAAF achieves scene-adaptive modality integration through two-stage progressive fusion; CMFE-SRP employs hierarchy-specific processing to balance spatial details and semantic enhancement; DS-STD increases positive sample coverage to 4× through spatial tolerance expansion. Experiments on VEDAI (satellite) and RGBT-Tiny (UAV) demonstrate mAP values of 62.22% and 18.52%, improving over the state of the art by 4.3% and 10.3%, with a 17.3% improvement on extremely tiny targets.

Förderung

  • Hunan Provincial Postgraduate Research Innovation Programme

Zitieren

APA

Zhou, P., Guo, X., Sun, X., Sun, B., Su, S., Jiang, W., Guo, R., Dang, Z., & Huang, S. (2026). RAPT-Net: Reliability-Aware Precision-Preserving Tolerance-Enhanced Network for Tiny Target Detection in Wide-Area Coverage Aerial Remote Sensing. Remote Sensing, 18(3), 449. https://doi.org/10.3390/rs18030449

Vancouver

Zhou P, Guo X, Sun X, Sun B, Su S, Jiang W, et al. RAPT-Net: Reliability-Aware Precision-Preserving Tolerance-Enhanced Network for Tiny Target Detection in Wide-Area Coverage Aerial Remote Sensing. Remote Sensing. 2026;18(3):449. doi:10.3390/rs18030449

BibTeX
@article{Zhou2026,
  author = {Zhou, Peida and Guo, Xiaojun and Sun, Xiaoyong and Sun, Bei and Su, Shaojing and Jiang, Wei and Guo, Runze and Dang, Zhaoyang and Huang, Siyang},
  title = {RAPT-Net: Reliability-Aware Precision-Preserving Tolerance-Enhanced Network for Tiny Target Detection in Wide-Area Coverage Aerial Remote Sensing},
  journal = {Remote Sensing},
  year = {2026},
  volume = {18},
  number = {3},
  pages = {449},
  doi = {10.3390/rs18030449}
}
Kostenlos starten →