.png){kind=icon}
.png){kind=icon}
Return
Global mapping of stony desert through few-shot similarity with geospatial foundation models
.png){kind=icon}
OA
AI
王
N
Y
L
J
DOI：10.1016/j.geoderma.2026.117982.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• GFMs support global stony desert mapping with few-shot similarity retrieval. • GFM embeddings improve SD classification over conventional multispectral data. • FSR approaches SDL performance with fewer reference samples. • The 2025 global 10 m stony desert map covers an area of 6.34 ± 0.15 million km2.
Keywords:
Stony desert
Geospatial foundation models
Few-shot learning
Cross-regional retrieval
Remote sensing classification
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.
Journal
IF:
6.6
Papers:
9.3K
Citations:
4.5W
Organization
Cited Papers