Computer Science · Computer Vision and Pattern Recognition
DOI: 10.1109/ICIP.2009.5413453 NON-LOCALIMAGESMOOTHINGBY APPLYING ANISOTROPICDIFFUSIONPDE’S INTHE SPACEOF PATCHES
See on the mapAbstract
Wedesignafamilyof non-local image smoothing algorithms which approximate the application of diffusion PDE’s on a specific Euclidean space of image patches. We first map a noisy image onto this high-dimensional space and estimate its geometric structure thanks to a straightforward extension of the structure tensor field. The tensors spectral elements allows us to design an oriented highdimensional smoothingprocess bythemeans ofanisotropic regularization PDE’s which have both local and non-local properties and whose solutions are estimated by locally oriented high-dimensional convolutions. We show that the Bilateral Filtering and Non-Local Means methods are the isotropic cases of our denoising framework. Index Terms — Edge-preserving denoising, non-local smoothing, anisotropic diffusion, regularizationPDE’s. 1.