A global sparse gradient based coupled system for image denoising

Abstract

We consider a coupled system for image denoising, which consists of an anisotropic diffusion equation and a global sparse gradient model (GSG). The global sparse gradient model is used to estimate the edge map from the noisy image accurately and robustly. The anisotropic diffusion process is guided by the edge map. Furthermore, we prove the existence and uniqueness of the solutions of the coupled system with Dirichlet initial–boundary value problem. Finally, comparing with the edge indicators based on local methods, experimental results demonstrate that our proposed method has the state-of-the-art performance among the PDE-based anisotropic diffusion approaches both in objective measurement and visual evaluation.