Identification of rice paddy fields from multitemporal polarimetric SAR images by scattering matrix decomposition

Rice is the staple food in many countries and monitoring the area of rice fields is demanded. The backscatter behavior of rice as function of time is so unique that it can be used to distinguish rice from other crops. In this paper, scattering matrix decomposition models were adopted to identify rice paddy fields from multi-temporal polarimetric SAR images. A series of Radarsat-2 polarimetric SAR images was acquired during the whole cultivation period over a rice site. Experimental results revealed that volume scattering mechanism is more sensitive to rice growth than other scattering mechanisms. In addition, rice fields can be successfully identified using volume scattering power no matter what decomposition model is applied.