Multiscale segmentation of SAR imagery with bootstrap sampling

In this paper, the main focus is on application of bootstrap method for segmentation of SAR imagery based on the multiscale dynamic model. Given an original SAR images, we randomly select a small representative set of pixels. Then, multiscale autoregressive (MAR) model, which captures multiscale stochastic structure inherent in SAR imagery, is used for SAR imagery segmentation. The results we obtain show that our method based on bootstrap sampling gives better results than the classical one both in the quality of the segmented image and the computational time.