SIMULATION AND ANALYSIS OF BACKSCATTERING CHARACTERISTICS AND TYPE RECOGNITION OF CROPS BASED ON DUAL-FREQUENCY AND MULTI-POLARIZATION SAR DATA

Abstract

Based on C-band ENVISAT/ASAR and L-band ALOS/PALSAR data obtained from Changping test site of Beijing in 2007,the authors used the MIMICS model to simulate and analyze the backscattering characteristics of spring maize,summer maize and fruit trees,then compared the simulation result with the backscattering coefficients extracted from the SAR image and,according to the difference in backscattering coefficients of crops,set up a binary tree of classification algorithm to discriminate and identify the spring maize and summer maize,with the overall classification accuracy up to 86.66%.The result indicates that the dual-frequency and multi-polarization data can provide multi-dimensional information for identification and classification of crops,and have significant potential and advantages for crop remote sensing.