EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Mapping Surface-Cover Components in Desert Steppe from UAV RGB Imagery Using Interpretable Features and Multiscale Superpixel Probability Fusion

Haiyang Wang, Yuge Bi, Sen Wang, Wenye Xie, Yingshuo Du, Zhanzhao Liu

Agronomy · 05.11.2026 · 16 (21) · 2213

DOI: 10.3390/agronomy16212213 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Agronomy, online erschienen am 05.11.2026. Verfasst von Haiyang Wang und 5 weiteren Autor:innen. Die Arbeit nennt 32 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Natural Science Foundation of Inner Mongolia Autonomous Region, Natural Science Key Research Project of Higher Education Institutions of Inner Mongolia Autonomous Region.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Unmanned aerial vehicle (UAV) red–green–blue (RGB) imagery provides low-cost, high-resolution observations of desert-steppe cover, but overlapping colors and spatial heterogeneity complicate classification. We evaluated random forest (RF) with multiscale superpixel probability fusion (RF-MSPF) using 19 patches from nine images and 335,675 reference pixels representing herbaceous vegetation, woody or semi-woody vegetation, visible dead plant material, and bare substrate. Five interpretable RGB features supplied RF class probabilities, aggregated within Simple Linear Iterative Clustering (SLIC) regions at three scales. Source-image-grouped five-fold validation increased mean overall accuracy (OA) from 0.8020 ± 0.0573 for RF to 0.8691 ± 0.0367 for RF-MSPF (mean ± sample standard deviation), with gains in every fold. Equal-weight fusion achieved 0.8704 ± 0.0367. On the fixed development split, RF-MSPF achieved OA = 0.8806, macro-averaged F1 (Macro-F1) = 0.8742, and mean intersection over union (mIoU) = 0.7840. Ablations identified regional aggregation as the main improvement, with further gains from multiscale fusion; consistency weighting added only 0.024 percentage points on that split and showed no consistent advantage. This transparent, reproducible workflow improved classification within this dataset.

Förderung

  • Natural Science Foundation of Inner Mongolia Autonomous Region
  • Natural Science Key Research Project of Higher Education Institutions of Inner Mongolia Autonomous Region

Zitieren

APA

Wang, H., Bi, Y., Wang, S., Xie, W., Du, Y., & Liu, Z. (2026). Mapping Surface-Cover Components in Desert Steppe from UAV RGB Imagery Using Interpretable Features and Multiscale Superpixel Probability Fusion. Agronomy, 16(21), 2213. https://doi.org/10.3390/agronomy16212213

Vancouver

Wang H, Bi Y, Wang S, Xie W, Du Y, Liu Z. Mapping Surface-Cover Components in Desert Steppe from UAV RGB Imagery Using Interpretable Features and Multiscale Superpixel Probability Fusion. Agronomy. 2026;16(21):2213. doi:10.3390/agronomy16212213

BibTeX
@article{Wang2026,
  author = {Wang, Haiyang and Bi, Yuge and Wang, Sen and Xie, Wenye and Du, Yingshuo and Liu, Zhanzhao},
  title = {Mapping Surface-Cover Components in Desert Steppe from UAV RGB Imagery Using Interpretable Features and Multiscale Superpixel Probability Fusion},
  journal = {Agronomy},
  year = {2026},
  volume = {16},
  number = {21},
  pages = {2213},
  doi = {10.3390/agronomy16212213}
}
Kostenlos starten →