Statistical Analysis of Typical Clutter Scenes in High Resolution SAR Images

Abstract

In the ground motion target indication (GMTI), the echo signal generated by the stationary target is called clutter. Statistical analysis of synthetic aperture radar (SAR) image clutter describes the SAR image clutter data by statistical methods, so as to reveal the statistical characteristics of clutter. The study of clutter statistical property lays the foundation for the research of clutter suppression algorithm and constant false alarm rate (CFAR) detection algorithm. It also helps understand the scattering mechanism of ground objects and provides assistance for the research of SAR image speckle suppression, segmentation, classification and RCS reconstruction. This paper focuses on the clutter statistical modeling method based on parametric model. The empirical distribution model, the multiplicative noise model and their parameter estimation methods are introduced at first. Then some goodness of fit test methods are summarized. Finally, the statistical analysis of clutter statistical modeling experiment based on the spaceborne SAR image data shows the statistical characteristics of typical clutter scenes.