The new generation of Synthetic Aperture Radar (RADARSAT-2, TerraSAR-X, ALOS, . . . ) allows us to capture Earth surface images with very high resolution. Therefore the possibility to characterize objects has become more and more attainable. Moreover, the short revisit time propertie of these satellites enables the development of techniques of change detection and their applications. Spherically Invariant Random Vector (SIRV) model was designed specifically for the analysis of heterogeneous clutters in high resolution radar images. In this paper, we propose four algorithms of change detection based on different criteria including: Gaussian (sample covariance matrix estimator), Gaussian (fixed point estimator), Fisher texture-based and KummerU-based (Fisher distributed texture).