Optimal Grid-Based Filtering for Crop Phenology Estimation with Sentinel-1 SAR Data

In the last decade, suboptimal Bayesian filtering (BF) techniques, such as Extended Kalman Filtering (EKF) and Particle (PF), have led to great interest for crop phenology monitoring with Synthetic Aperture Radar (SAR) data. this study, a novel approach, based on Grid-Based Filter (GBF), is proposed estimate phenology. Here, phenological scales, which consist of finite number discrete stages, represent one-dimensional state space, hence GBF provides optimal estimates. Accordingly, contrarily literature studies EKF PF, no constraints are imposed models statistical distributions involved. The prediction model defined by transition matrix, while Kernel Density Estimation (KDE) employed define observation model. approach applied dense time series dual-polarization Sentinel-1 (S1) SAR images, collected in four different years, BBCH stages rice crops. Results show that 0.94 ≤ R2 0.98, 5.37 RMSE 7.9 20 MAE 33.