EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

DCA-DeepLab: Dual-Coordinate Attention DeepLab with Adaptive Focal Loss for Cotton Growth Semantic Segmentation from UAV Remote Sensing Images

Liruizhi Jia, Jiazhan Gao, Zuolong Li, Heng Shi, Jihong Zhu

Drones · 11.06.2026 · 10 (6) · 456

DOI: 10.3390/drones10060456 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Drones, online erschienen am 11.06.2026. Verfasst von Liruizhi Jia und 4 weiteren Autor:innen. Die Arbeit nennt 76 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Open Foundation of the State Key Laboratory of Precision Space-time Information Sensing Technology, Innovation Program for Doctoral Students of Xinjiang University.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

UAV remote sensing provides centimetre-level imagery for fine-grained cotton growth monitoring, yet existing segmentation models face three challenges: cotton fields exhibit a pronounced row and column structure that standard convolutions struggle to capture; conventional decoders fuse features statically, suppressing fine boundary cues; and the pixel-level class distribution is severely imbalanced. We present DCA-DeepLab, built on DeepLabv3+ with three task-specific components: a Dual-Coordinate Attention Gating (DCAG) module that decouples horizontal and vertical dependencies to encode row and column structures; a Multi-Scale Attention-Guided Modulated Feature Merging (MSAM-MFM) module that reweights semantic and detail features at each location; and an adaptive pixel-level modulated focal loss (APMFL), which focuses training on hard, minority-class pixels. We construct a cotton growth dataset of 11,745 UAV patches with four semantic classes. On this dataset and the public LoveDA benchmark, DCA-DeepLab attained the highest mIoU among the compared methods (51.74% and 51.71%), exceeding the strongest cotton baseline by 1.10 percentage points. Relative to DeepLabv3+, the Vigorous and Sparse minority-class IoUs improved by 3.51 and 1.91 percentage points, respectively, and Vigorous recall rose from 51.85% to 60.04%, with only 3.9% more parameters. These results show that encoding directional structure and adaptively balancing class contributions benefits fine-grained UAV crop segmentation.

Förderung

  • Open Foundation of the State Key Laboratory of Precision Space-time Information Sensing Technology
  • Innovation Program for Doctoral Students of Xinjiang University

Zitieren

APA

Jia, L., Gao, J., Li, Z., Shi, H., & Zhu, J. (2026). DCA-DeepLab: Dual-Coordinate Attention DeepLab with Adaptive Focal Loss for Cotton Growth Semantic Segmentation from UAV Remote Sensing Images. Drones, 10(6), 456. https://doi.org/10.3390/drones10060456

Vancouver

Jia L, Gao J, Li Z, Shi H, Zhu J. DCA-DeepLab: Dual-Coordinate Attention DeepLab with Adaptive Focal Loss for Cotton Growth Semantic Segmentation from UAV Remote Sensing Images. Drones. 2026;10(6):456. doi:10.3390/drones10060456

BibTeX
@article{Jia2026,
  author = {Jia, Liruizhi and Gao, Jiazhan and Li, Zuolong and Shi, Heng and Zhu, Jihong},
  title = {DCA-DeepLab: Dual-Coordinate Attention DeepLab with Adaptive Focal Loss for Cotton Growth Semantic Segmentation from UAV Remote Sensing Images},
  journal = {Drones},
  year = {2026},
  volume = {10},
  number = {6},
  pages = {456},
  doi = {10.3390/drones10060456}
}
Kostenlos starten →