SIAM Digital Library
 
 
 

You are not logged in Logged Out Log In

Multiscale Model. Simul. 4, pp. 584-609 (26 pages)

Image Registration Based on Multiscale Energy Information

Stefan Henn and Kristian Witsch

Full Text: Download PDF | Buy PDF (US$25) | View Cart
We propose a novel multiscale approach for the image registration problem, i.e., to find a deformation that maps one image onto another. The image registration problem is confirmed to be mathematical ill-posed due to the fact that determining the unknown components of the displacements merely from the images is an underdetermined problem. The approach presented here utilizes an auxiliary regularization term based on the energy of a plate with free edges, which incorporates smoothness constraints into the deformation field. One of the important aspects of this approach is that the energy does not penalize affine-linear functions. Consequently, the kernel of the Euler--Lagrange equation is spanned by all rigid motions. Hence, the presented approach is invariant under planar rotation and translation. In order to find an optimal deformation, we solve a sequence of subproblems with decreasing regularization parameter. In this framework the regularization parameter can be regarded as a scale parameter, which captures information at multiple spatial scales. We analyze the multiscale nature of a solution.

© 2005 Society for Industrial and Applied Mathematics

RELATED DATABASES

To view database links for this article, you need to log in.

PUBLICATION DATA

ISSN:

1540-3459 (print)  
1540-3467 (online)

ARTICLE DATA


For access to fully linked references, you need to log in.

For access to citing articles, you need to log in.


Close

close