An Unsupervised Classification for Fully Polarimetric SAR Data Using Cloude-Pottier Decomposition and Agglomerative Hierarchical Clustering Algorithm

An unsupervised classification method is proposed for fully polarimetric SAR data.The total backscattering power SPAN combined with the entropy H,the α angle and the anisotropy A is used to initialize the polarimetric SAR data.An agglomerative hierarchical clustering algorithm is introduced to reduce the number of clusters.The experimental results show that the SPAN has the additional information that is not contained in the H/α/A,and this information is important for the initialization.It is also shown that the proposed classification algorithm provides better performance than the general Wishart H/α/A classification.