Synthetic Aperture Radar Image Statistical Modeling: Part Two-Spatial Correlation Models and Simulation

Following on the first part of our review of synthetic aperture radar (SAR) image statistical modeling [1], which concerns single-pixel statistical models, this article extends our discussion to spatial correlation analysis, focusing on SAR spatial correlation models and SAR clutter simulation methods. Two types of spatial correlation models, the product model and the coherent scatterer model, are summarized: the first considers correlation characteristics from the image itself while the second analyzes spatial correlation stemming from the underlying scatterer and the physical imaging process. In addition, we review four spatially correlated clutter simulation methods based on two classical distributions (< italic > K and < italic > G < sup > 0 ): a product model-based method, an inverse transform method (ITM)-based approach, a coherent scatterer model-based technique, and the generalized Gaussian coherent scatterer (< italic > GGCS ) approach. We discuss the advantages and disadvantages of these models and methods and provide references for further research into the statistical modeling of SAR images.

Automated summary

This article extends the discussion on SAR image statistical modeling to spatial correlation analysis, focusing on SAR spatial correlation models and clutter simulation methods. It summarizes two types of spatial correlation models and reviews four clutter simulation methods based on different distributions, providing references for further research in this area.