A Dynamical System Approach for Agricultural Crop Phenology Tracking using PolSAR Time-Series: Cereals Case Study at C-Band

Abstract:
In this work an approach to estimate the phenological stage of cereal crops with C-band radar data is proposed. The objective is to define a general methodology to estimate the current phenological stage of a crop during its cultivation cycle, exploiting time series of polarimetric SAR data. The approach is divided into two main steps: the definition of an evolution model and the estimation based on filtering techniques. An orthogonal transformation based on Principal Component Analysis (PCA) over the PolSAR observables series is used to define the state space of the dynamical process. The generation of the evolution model is carried out using a set of reference crop fields for which ground truth is available. The estimation approach employs a local linearization from a geometrical model which feeds an Extended Kalman Filter (EKF).