A dual algorithm for denoising and preserving edges in image processing
Philippe Destuynder, Samir Jaoua, Hela Sellami · Journal of Inverse and Ill-Posed Problems · 2007
The restoration is a primordial step on image processing. We propose a variational model which permits to simultanuously denoise and preserve edges on image. By the use of the duality theory we prove that the minimization problem can be solved by considering its dual. First we treat an existance and unicity results of solution of the dual problem, and then we describe the algorithm for computing the solution. Finally, we present some numerical results for images on one and two dimension.