Kernel Wiener filtering model with low-rank approximation for image denoising.

•The kernel Wiener filtering model is proposed as an extension of the conventional Wiener filtering by introducing a shape-aware kernel function.•The constrained covariance matrix nuclear norm is minimized to estimate the reference images.•Eigenvalue thresholding is deduced for the shrinkage of transformed coefficients in the optimized low-rank approximation.•The experiments verify the generality and effectiveness of the proposed method.