A novel SAR image denoising method based on self-supervised network

Noise suppression is an important preprocessing step for sythetic aperture radar (SAR) image interpretation. In particular, noise affects the quality of SAR images, which in turn affects the subsequent down streaming processing, such as target identification, classificatio, etc. In this paper, a novel SAR image denoising method based on self-supervised framework is proposed, which utilizes the statistical distribution characteristics of noise and conducts denoising processing through the self-supervised network framework, and then achieves high-quality denoised radar image. Moreover, the effectiveness of the proposed method is verified through computer simulation experiments.