We show in this paper that PolInSAR images can be efficiently partitioned into homogeneous regions with a statistical technique based on the minimization of a criterion without parameter to be tuned by the user. This technique consists in finding a polygonal partition of the image that minimizes the stochastic complexity assuming that the image is made of a tessellation of statistically homogeneous regions. The obtained results demonstrate that global partition in statistically homogeneous regions of PolInSAR images can provide better results than partition based on a single characteristic such as polarimetry or interferometry only. In particular, it enables us to obtain a good estimate of the PolInSAR covariance matrices for parameter extraction.