A speckle suppression SAR imaging method based on compressive sensing

Abstract

Speckle of synthetic aperture radar (SAR) image is the inherent defect of all types of coherent imaging systems. Usually, speckle suppression is discussed using data at or above Nyquist sampling rate. In this paper, we propose a speck suppression method for SAR image only using a small amount of data. For this purpose we transform the multiplicative noise into additive noise at first to fit the image model, and then compressive sensing (CS) theory is applied to reconstruct the image. The experimental results on synthetic image and real SAR image show that our method outperforms the basic CS reconstruction method.