Improved optimization methods for image registration problems

Chen, K.;Nunes Grapiglia, Geovani;Yuan, J.;Zhang, D.
(2019) Numerical Algorithms — Vol. 80, p. 305-336 (2019)

Files

proof.pdf
  • Open Access
  • Adobe PDF
  • 5.21 MB
Download

Details

Authors
  • Chen, K.University of Liverpool
    Author
  • Nunes Grapiglia, Geovaniorcid-logoUCLouvain
    Author
  • Yuan, J.UFPR
    Author
  • Zhang, D.University of Liverpool
    Author
Abstract
(en) In this paper, we propose new multilevel optimization methods for minimizing continuously differentiable functions obtained by discretizing models for image registration problems. These multilevel schemes rely on a novel two-step Gauss-Newton method, in which a second step is computed within each iteration by minimizing a quadratic approximation of the objective function over a certain two-dimensional subspace. Numerical results on image registration problems show that the proposed methods can outperform the standard multilevel Gauss-Newton method.
Affiliations

Citations

  • APA
  • Chicago
  • FWB

Chen, K., Nunes Grapiglia, G., Yuan, J., & Zhang, D. (2019). Improved optimization methods for image registration problems. Numerical Algorithms, 80, 305-336. https://doi.org/10.1007/s11075-018-0486-2 (Original work published 2019)