Assessing the value of advanced Sentinel-1 SLC features for multimodal time-series crop mapping
Year: 2026
Type: article
Abstract: • Extension of PASTIS-R dataset with Sentinel-1 SLC data for multimodal agricultural mapping. • Reflection symmetry of agricultural targets makes dual-pol decomposition features redundant with backscatter coefficients. • Interferometric coherence boosts robustness under severe optical cloud cover. • In multimodal fusion, biased coherence matches elaborately processed alternatives, simplifying operational deployment. • Coherence gains transfer across held-out regions, with larger marginal contributions under regional distribution shifts.
Institutions: PLA Information Engineering University, Duke University
Language: English
Subfield: Aerospace Engineering
Field: Engineering
Domain: Physical Sciences
SDG: Zero hunger
Open Access status: gold
Estimated APC paid: $3,550
OpenAlex ID: W7162095794