SSIM-based optimal non-local means image denoising with improved weighted Kernel function

In this paper, a novel SSIM-based NLM algorithm is proposed. We replace the exponential weighted kernel function with a cosine coefficient weighted Gaussian kernel function and incorporate the structural similarity (SSIM) index, which is the state-of-the-art image quality assessment measure, into the NLM image denoising model. Experimental results indicate that the proposed SSIM-based optimal NLM algorithm can get higher PSNR and SSIM, meanwhile, it keeps more detail information and structure feature of the original image and acts out the best visual quality.