ANALYSIS AND VERIFICATION OF THE CHARACTERISTICS OF THE SHARP-PEAKED AND HEAVY-TAILED OF GF-3 SAR IMAGE
Yunjing Song, Zengguo Sun · 2018
GF-3 is the first Chinese civil c-band synthetic aperture radar (SAR) launched on 10 August 2016 by China Academy of Space Technology, which operates in 12 imaging modes with a fine spatial resolution up to 1 m. The SAR images produced by the GF-3 satellite are rich in information and high-resolution. The study of the statistical distribution of SAR images has always been an important part of SAR image processing. In order to obtain high quality noise reduction and target detection results, it is necessary to accurately describe the statistical characteristics of GF-3 SAR images. During the statistical analysis of GF-3 SAR images, it was found that GF-3 SAR images have sharp-peaked and heavy-tailed statistical properties. Distributions of SAR images based on central limit theorem such as Rayleigh distribution cannot reflect the sharp-peak and heavy-tail of high-resolution GF-3 SAR images. Therefore, the sharp-peaked and heavy-tailed statistical properties of the heavy-tailed Rayleigh distribution is used to model the high-resolution GF-3 SAR images. In this paper, an example of GF-3 SAR images modeling based on heavy-tailed Rayleigh distribution is given. Modeling experiments on GF-3 SAR images show that, the heavy-tailed Rayleigh distribution accurately describes the sharp-parked and heavy-tailed statistical characteristics of the high resolution GF-3 SAR image when compared to the conventional Rayleigh, Weibull, and K distributions. The sharp-peaked and heavy-tailed impulse characteristics of the GF-3 image were verified by modeling the GF-3 SAR images using the heavy-tailed Rayleigh distribution.