English

The L1-Potts functional for robust jump-sparse reconstruction

Optimization and Control 2015-03-03 v3 Numerical Analysis Numerical Analysis

Abstract

We investigate the non-smooth and non-convex L1L^1-Potts functional in discrete and continuous time. We show Γ\Gamma-convergence of discrete L1L^1-Potts functionals towards their continuous counterpart and obtain a convergence statement for the corresponding minimizers as the discretization gets finer. For the discrete L1L^1-Potts problem, we introduce an O(n2)O(n^2) time and O(n)O(n) space algorithm to compute an exact minimizer. We apply L1L^1-Potts minimization to the problem of recovering piecewise constant signals from noisy measurements f.f. It turns out that the L1L^1-Potts functional has a quite interesting blind deconvolution property. In fact, we show that mildly blurred jump-sparse signals are reconstructed by minimizing the L1L^1-Potts functional. Furthermore, for strongly blurred signals and known blurring operator, we derive an iterative reconstruction algorithm.

Cite

@article{arxiv.1207.4642,
  title  = {The L1-Potts functional for robust jump-sparse reconstruction},
  author = {Andreas Weinmann and Martin Storath and Laurent Demaret},
  journal= {arXiv preprint arXiv:1207.4642},
  year   = {2015}
}
R2 v1 2026-06-21T21:38:25.598Z