Image Denoising Algorithm Based on Gradient Domain Guided Filtering and NSST

Traditional image denoising methods, which do not depend on data training, have good interpretability. However, traditional methods hardly achieve the effect of deep learning methods. Based processing techniques, this paper proposes a new hybrid model. The block-batching and 3-D filtering (BM3D) algorithm is used to obtain first denoised image. weighted kernel norm minimization (WNNM) non-subsampled shearlet transform (NSST) algorithms are successively adopted get second By gradient domain guided filtering, texture information extracted enhance details Specially, we propose adaptive iterative NSST based improved soft thresholding, in order solve problems about discontinuity hard thresholding constant deviation thresholding. Our approach can only attenuate excessive smoothing, but also restore natural appearance Experiments conducted demonstrate that our proposed method enjoys PSNR SSIM performance gains over several