.png){kind=icon}
.png){kind=icon}
Return
High-resolution forest height retrieval from L-band interferometric SAR time series using deep learning over Northern Spain
.png){kind=icon}
OA
AI
C
O
M
DOI：10.1016/j.srs.2026.100457.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• ALOS-2 PALSAR-2 dual-pol InSAR time series enabled accurate forest height retrieval. • Physics-model based inversion was coupled with U-Net modeling of L-band InSAR data. • Achieved 2.0–2.8 m RMSE, outperforming semi-empirical and machine learning baselines. • Attention mechanisms and nested feature fusion provide better prediction accuracy. • Aggregation to 40–60 m resolution balanced accuracy and detail for scalable mapping.
Keywords:
ALOS-2 PALSAR-2
Synthetic aperture radar
L-band
Polarimetry
Interferometry
Deep learning
Forest mapping
Regression modeling
Journal
IF:
5.2
Papers:
457
Citations:
980
Organization
Cited Papers