Nonlocal Operators with Applications in Imaging

Stanley Osher
University of California, Los Angeles (UCLA)

We will review the use of nonlocal operators to define nonlocal flows and variational methods arising in image processing and elsewhere. This is based on nonlocal means devised by Buades, Coll, Morel and graph based gradients and divergences devised in
machine learning.
This has proven to be quite useful in imaging, but we hope to extend its utility to high dimensional Hamilton-Jacobi equations arising in control,and elsewhere.


Back to Nonlocal PDEs, Variational Problems and their Applications