Curvature regularization of image level lines is a powerful tool in image processing. Using so-called functional lifting, this can be achieved by specific convex functionals in a higher-dimensional space. The functional requires a subtle discretization of a Radon measure to fulfill a compatibility condition and to give reasonable results. Additionally, the resulting high computational costs have to be managed. We present an adaptive discretization and give some results for image segmentation for 2D- and 3D-images.

This presentation is part of Minisymposium “MS31 - Variational Approaches for Regularizing Nonlinear Geometric Data (3 parts)”
organized by: Martin Storath (Universität Heidelberg) , Martin Holler (École Polytechnique, Université Paris Saclay) , Andreas Weinmann (Hochschule Darmstadt) .

Authors:

Ulrich Hartleif (Universität Münster)

Keywords:

adaptive discretization, functional lifting, image segmentation