Multiscale models for target detection and background discrimination in synthetic aperture radar imagery

Multiscale models can be used to capture the scale-dependent behavior of th e statistics in radar imagery. This behavior is expected to be different fo r natural background compared to objects of interest such as vehicles. We d emonstrate that multiscale autoregressive models can discriminate between s amples of these two major classes extracted from 1.5-m-resolution radar ima gery. We also show that it is possible to discriminate between two types of natural background in SAR imagery, "grassland" and "woodland," using multi scale models. This latter result could be exploited in adaptive algorithms for automated target detection. (C) 1999 Academic Press.