PolSAR change detection for specific land cover type by testing equality of two PolInSAR coherency matrixes

In this paper, we will propose a novel polarimetric SAR (PoISAR) change detection method applied to specific land cover type. Firstly, a polarimetric SAR interferometry (PolInSAR) coherency matrix is used to simultaneously take into account the full polarimetric information from both images. Then, a generalized likelihood ratio test (GLRT) statistic for equality of two polarimetric coherency matrixes is considered and a maximum likelihood distance measure is presented to extract the difference map of the change from specific class w_j, to specific class. Afterwards, Kittler and Illingworth (KI) minimum error threshold segmentation method is applied to extract the specific changed areas. Two Radarsat-2 fully polarimetric images in Suzhou city, China, acquired on April 9, 2009 and June 15, 2010 separately, are used for our experiment. It is shown that the proposed change detection method will give a good performance to achieve the specified change areas of PoSAR images.