Benchmark RadarAI BENCHMARK PROFILE
SwissCrop25
Evaluates crop mapping models on national-scale multi-year Sentinel-2 data with fine-grained crop taxonomy and non-crop classes under leave-one-year-out protocol.
- Released
- 2026-08-10
- Readiness
- Inspectable
- Primary field
- General AI
Why it matters
Provides a realistic operational testbed that exposes interannual distribution shifts and architecture differences hidden by conventional benchmarks.
Motivation
Operational crop mapping requires models that generalise across years, resolve fine-grained crop taxonomies, and distinguish cropland from surrounding landscapes.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.