Affordable Access

Download Read

Applications of nonlinear diffusion in image processing and computer vision

Authors
Keywords
  • 004 Informatik

Abstract

Nonlinear diffusion processes can be found in many recent methods for image processing and computer vision. In this article, four applications are surveyed: nonlinear diffusion filtering, variational image regularization, optic flow estimation, and geodesic active contours. For each of these techniques we explain the main ideas, discuss theoretical properties and present an appropriate numerical scheme. The numerical schemes are based on additive operator splittings (AOS). In contrast to traditional multiplicative splittings such as ADI, LOD or D'yakonov splittings, all axes are treated in the same manner, and additional possibilities for efficient realizations on parallel and distributed architectures appear. Geodesic active contours lead to equations that resemble mean curvature motion. For this application, a novel AOS scheme is presented that uses harmonie averaging and does not require reinitializations of the distance function in each iteration step.

There are no comments yet on this publication. Be the first to share your thoughts.

Statistics

Seen <100 times
Downloaded <100 times
0 Comments

More articles like this

A new nonlocal nonlinear diffusion of image proces...

on Journal of Differential Equati... Jan 01, 2009

Tensor scale: A local morphometric parameter with...

on Computer Vision and Image Unde... Jan 01, 2005
More articles like this..