
Chapter 4
Nonlocal Variational Methods in
Image Restoration
The traditional regularization terms previously discussed (depending on the
local image derivatives) are based on local image operators (like the image
gradient), which denoise a nd preserve edges very well, but may induce loss
of fine structures like texture during the restoratio n process. More recently,
Buades et al. [72] introduced the nonlocal means filter or NL–means (as a
typ e of neighborhood filter), producing excellent denoising results. Gilboa
and Osher [154, 155] formulated the variational framework of NL–means by
proposing nonlo cal regularizing functionals using nonlocal operators ...