Label-free inference of latent intra-annual multi-cropping sequences from multi-sensor SITS: Fourier-basis gap reconstruction, batched bounded LM double-logistic inversion, α-threshold phenometrics, rule-induced pseudo-labels, spatially-blocked RF/TempCNN, MK–Theil–Sen operators, FAO-56 ET accounting
random-forest evapotranspiration weak-supervision levenberg-marquardt sits sar google-earth-engine sentinel-2 phenometrics sentinel-1 pseudo-labelling land-surface-phenology harmonic-regression satellite-image-time-series spatial-cross-validation spatiotemporal-learning tempcnn double-logistic multi-cropping crop-sequence-inference
-
Updated
Nov 6, 2026 - Python