Image super-resolution by adaptive coupling of oriented diffusion and shock filter

We present a new single image super-resolution approach based on Partial Differential Equations (PDE), which consists of oriented diffusion, fidelity term and shock filter. It weakens the blurred image and keeps its characters by fidelity coefficient related with edges, weighs diffusion and shock filtering through continuous function based on content of image, adaptively combine the two kinds of PDE. Experiments demonstrate that the proposed adaptive coupling approach can achieve better restoration result than traditional interpolation methods and other PDE-based methods.