Change detection based on polarization decomposition using RADARSAT-2 Quad-pol data

Thanks to the capability to operate in almost all weather conditions and during both day and night time, change detection (CD) based on SAR data is developed rapidly in recent years, especially with the successful operation of full polarization space-borne SAR system. Most of the CD methods based on Quad-pol SAR data are through the analysis of statistical characteristics of the polarimetric covariance matrix or coherency matrix. In this study, we proposed a new CD method by comparing the difference between the scatter components, obtained by polarimetric target decomposition, to extract the difference map and then an adaptive KI algorithm is carried out to find an appreciate threshold for segmenting to get the change. Two RADARSAT-2 quad polarimetric images acquired over the Suzhou City in China are analyzed for validation in the experiment and the effectiveness of the proposed method is presented in the result.