SAR Images Denoising Using Bidimensional Variational Mode Decomposition and Nonlocal Means Reprojection with Minimizing Variance
Harouchi Badre, Yassine Tounsi, Mohammed Said Rachafi, Hamid Bioud, Abdelkrim Nassim · Geosciences · 2021
Synthetic aperture radar (SAR) is very high and accurate technology for remote sensing. SAR images are very exploited for change detection and monitoring because of their high resolution. Because of the coherent retro diffusion of the backscattered radar waves with the rigorous earth surface, SAR images are characterized by a multiplicative noise called speckle that influences their analysis. For this reason, this paper aims to exploit the two-dimensional variational mode decomposition and combines it with the nonlocal means reprojection method to denoise SAR images. The proposed method concerns to decompose the SAR image into a series of variational mode functions called S-D IMFs and N-D IMFs, however, the speckle noise is high-frequency information, which means that it is concentrated in N-D IMFs components. Then, the denoising of N-D IMFs components is implemented using NLM reprojection using box kernel with minimizing variance. Firstly, we use numerical simulation to study the performance of our method, this study concerns use several synthetic images and add speckle noise for different variance, and then we apply the proposed method. The quantitative appraisal will be realized using three important metrics such as peak signal to noise ratio (PSNR), Edge preservation index (EPI), and image quality index (Q). Also, we compare our proposed method with other famous techniques as the Lee filter, Frost filter, and NLM-SAR method. Secondly, the proposed method will be exploited to denoise a real SAR image.