Abstract
Image registration, i.e., finding an optimal displacement field u which minimizes a distance functional D(u) is known to be an ill-posed problem. In this paper a novel variational image registration method is presented, which matches two images acquired from the same or from different medical imaging modalities. The approach proposed here is also independent of the image dimension. The proposed variational penalty against oscillations in the solutions is the standard H2(Ω) Sobolev semi-inner product for each component of the displacement. We investigate the associated Euler-Lagrange equation of the energy functional. Furthermore, we approach the solution of the underlying system of biharmonic differential equations with higher order boundary conditions as the steady-state solution of a parabolic partial differential equation (PDE).
One of the important aspects of this approach is that the kernel of the Euler-Lagrange equation is spanned by all rigid motions. Hence, the presented approach includes a rigid alignment. Experimental results on both synthetic and real images are presented to illustrate the capabilities of the proposed approach.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
Y. Amit, “A nonlinear variational problem for image matching,” SIAM J. Sci. Comput. Vol. 15, pp. 207–224, 1994.
R. Bajcsy and S. Kovacic, “Multiresolution elastic matching,” Computer Vision, Vol. 46, pp. 1–21, 1989.
A. Barry, “Seeking signs of intelligence in the theory of control,” SIAM NEWS, Vol. 30, No. 3, 1997.
G. Christensen, M. Miller, M. Vannier, and U. Grenander, “Individualizing neuroanatomical atlases using a massively parallel computer,” IEEE Computer, Vol. 29, No. 1, pp. 32–38, 1996.
U. Clarenz, M. Droske, S. Henn, M. Rumpf, and K. Witsch, “Computational methods for nonlinear image registration,” 2005 to appear.
M. Davis, A. Khotanzad, D. Flaming, and S. Harms, “A physics based coordinate transformation for 3D medical images,” IEEE Trans. on medical imaging, Vol. 16, No. 3, pp. 317–328, 1997.
B. Fischer and J. Modersitzki, “Curvature based image registration,” JMIV, Vol. 18, pp. 81–85, 2003.
W. Hackbusch, Elliptic Differential Equations. Theory and Numerical Treatment. Springer Series in Computational Mathematics 18. Springer-Verlag, Berlin Heidelberg New York.
S. Henn, Numerische Lösung und Modellierung eines inversen Problems zur Assimilation digitaler Bilddaten, Phd thesis Heinrich-Heine-Universität Düsseldorf, Logos-Verlag, Berlin, 2001.
S. Henn and K. Witsch, “Iterative Multigrid Regularization Techniques For Image Matching,” SIAM J. Sci. Comput. (SISC), Vol. 23, No. 4, pp. 1077–1093, 2001.
S. Henn and K. Witsch, “Multi-Modal Image Registration Using A Variational Approach,” SIAM J. Sci. Comput. (SISC), Vol. 25, No. 4, pp. 1429–1447, 2004.
S. Henn and K. Witsch, “A variational image registration approach based on curvature scale space,” in Proceedings of the 5th International Conference on Scale Space and PDE Methods in Computer Vision 2005, pp. 143–154, 2005.
G. Hermosillo, “Variational methods for multimodal image matching,” Phd thesis, Université de Nice, France, 2002.
J. Maintz and M. Viergever, “A survey of medical image registration,” Medical Image Analysis, Vol. 2, No. 1, pp. 1–36, 1998.
J. Modersitzki, Numerical Methods for Image Registration, Oxford Science Publications. 2003.
P. Thompson and A. Toga, “Anatomically driven strategies for high-dimensional brain image registration and pathology,” Brain Warping, Academic Press, pp. 311–336, 1998.
Author information
Authors and Affiliations
Corresponding author
Additional information
Stefan Henn obtained his diploma (1997) and his Ph.D. in mathematics (2001), both from the Heinrich-Heine University (HHU) of Düsseldorf (Germany). From 1997–1999 he had a researcher position at the Institute for Brain Research at the HHU Düsseldorf. Since 1999 he is a research assistant at the Institute of Mathematics at the HHU Düsseldorf. He received the SIAM outstanding paper prize in 2003 for the paper (Iterative Multigrid Regularization Techniques for Image Matching, SIAM Journal on Scientific Computing, 23(4), pp. 1077-1093). His research interests include Multiscale methods in Scientific Computing and Image Processing, nonlinear large-scale optimization, and numerical analysis of partial differential equations.
Rights and permissions
About this article
Cite this article
Henn, S. A Full Curvature Based Algorithm for Image Registration. J Math Imaging Vis 24, 195–208 (2006). https://doi.org/10.1007/s10851-005-3621-3
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10851-005-3621-3