A Splitting Method for Diffeomorphism Optimization Problem Using Beltrami Coefficients

详细信息    查看全文

• 作者：Lok Ming Lui ; Tsz Ching Ng
• 关键词：Beltrami holomorphic flow ; Diffeomorphism optimization problem ; Beltrami coefficient ; Quasi ; conformal theories ; Alternating direction method of multiplier
• 刊名：Journal of Scientific Computing
• 出版年：2015
• 出版时间：May 2015
• 年：2015
• 卷：63
• 期：2
• 页码：573-611
• 全文大小：12,125 KB
• 参考文献：1. Lui, LM, Wong, TW, Zeng, W, Gu, XF, Thompson, PM, Chan, TF, Yau, ST (2012) Optimization of surface registrations using Beltrami holomorphic flow. J. Sci. Comput. 50: pp. 557-585 CrossRef
  2. Angenent, S, Haker, S, Tannenbaum, A, Kikinis, R (1999) On the Laplace–Beltrami operator and brain surface flattening. IEEE Trans. Med. Imaging 18: pp. 700-711 CrossRef
  3. Durrleman, S., Pennec, X., Trouve, A., Thompson, P., Ayache, N.: Measuring brain variability via sulcal lines registration: a diffeomorphic approach. In: Medical Image Computing and Computer-Assisted Intervention (MICCAI 2007) Lecture Notes in Computer Science 4791, pp. 675-82 Springer, Berlin, Heidelberg (2007)
  4. Durrleman, S, Pennec, X, Trouve, A, Thompson, P, Ayache, N (2008) Inferring brain variability from diffeomorphic deformations of currents: an integrative approach. Med. Image Anal. 12: pp. 626-637 j.media.2008.06.010" target="_blank" title="It opens in new window">CrossRef
  5. Fischl, B, Sereno, MI, Tootell, RB, Dale, AM (1999) High-resolution intersubject averaging and a coordinate system for the cortical surface. Hum. Brain Mapp. 8: pp. 272-284 CrossRef
  6. Gardiner, F (2000) Quasiconformal Teichmüller Theory. American Mathematical Society, Providence
  7. Glaunès, J, Vaillant, M, Miller, MI (2004) Landmark matching via large deformation diffeomorphisms on the sphere. J. Math. Imaging Vis. 20: pp. 179-200 CrossRef
  8. Gu, X, Wang, Y, Chan, T, Thompson, P, Yau, S-T (2004) Genus zero surface conformal mapping and its application to brain surface mapping. IEEE Trans. Med. Imaging 23: pp. 949-958 CrossRef
  9. Haker, S, Angenent, S, Tannenbaum, A, Kikinis, R, Sapiro, G, Halle, M (2000) Conformal surface parameterization for texture mapping. IEEE Trans. Vis. Comput. Graph. 6: pp. 181189
  10. Hurdal, M, Stephenson, K (2004) Cortical cartography using the discrete conformal approach of circle packings. NeuroImage 23: pp. S119S128 j.neuroimage.2004.07.018" target="_blank" title="It opens in new window">CrossRef
  11. Hurdal, MK, Stephenson, K (2009) Discrete conformal methods for cortical brain flattening. Neuroimage 45: pp. 86-98 j.neuroimage.2008.10.045" target="_blank" title="It opens in new window">CrossRef
  12. Joshi, S, Miller, M (2000) Landmark matching via large deformation diffeomorphisms. IEEE Trans. Image Process. 9: pp. 13571370 CrossRef
  13. Ju, L., Stern, J., Rehm, K., Schaper, K., Hurdal, M., Rottenberg, D.: Cortical surface flattening using least square conformal mapping with minimal metric distortion. In: IEEE International Symposium on Biomedical, Imaging pp. 77-0 (2004)
  14. Lehto, O, Virtanen, K (1973) Quasiconformal Conformal Mappings in the Plane. Springer, New York CrossRef
  15. Gardiner, F, Lakic, N (2000) Quasiconformal Teichmuller Theory. American Mathematical Society, Providence
  16. Leow, A, Yu, C, Lee, S, Huang, S, Protas, H, Nicolson, R, Hayashi, K, Toga, A, Thompson, P (2005) Brain structural mapping using a novel hybrid implicit/explicit framework based on the level-set method. NeuroImage 24: pp. 910-927 j.neuroimage.2004.09.022" target="_blank" title="It opens in new window">CrossRef
  17. Lepore, N., Leow, P.T.A.D. : Landmark matching on the sphere using distance functions. In: IEEE International Symposium on Biomedical Imaging (ISBI2006), April 6- 2006.
  18. Lord, NA, Ho, J, Vemuri, B, Eisenschenk, S (2007) Simultaneous registration and parcellation of bilateral hippocampal surface pairs for local asymmetry quantification. IEEE Trans. Med. Imaging 26: pp. 471-478 CrossRef
  19. Lui, L, Wang, Y, Chan, T, Thompson, P (2007) Brain anatomical feature detection by solving partial differential equations on general manifolds. Discrete Contin
• 刊物类别：Mathematics and Statistics
• 刊物主题：Mathematics
  Algorithms
  Computational Mathematics and Numerical Analysis
  Applied Mathematics and Computational Methods of Engineering
  Mathematical and Computational Physics
  
• 出版者：Springer Netherlands
• ISSN：1573-7691

文摘

Finding a meaningful 1- correspondence between different data, such as images or surface data, has important applications in various fields. It involves the optimization of certain energy functionals over the space of all diffeomorphisms. This type of optimization problems (called the diffeomorphism optimization problems, DOPs) is especially challenging, since the bijectivity of the mapping has to be ensured. Recently, a method, called the Beltrami holomorphic flow (BHF), has been proposed to solve the DOP using quasi-conformal theories (Lui et al. in J Sci Comput 50(3):557-85, 2012). The optimization problem is formulated over the space of Beltrami coefficients (BCs), instead of the space of all diffeomorphisms. BHF iteratively finds a sequence of BCs associated with a sequence of diffeomorphisms, using the gradient descent method, to minimize the energy functional. The use of BCs effectively controls the smoothness and bijectivity of the mapping, and hence makes it easier to handle the constrained optimization problem. However, the algorithm is computationally expensive. In this paper, we propose an efficient splitting algorithm, based on the classical alternating direction method of multiplier (ADMM), to solve the DOP. The basic idea is to split the energy functional into two energy terms: one involves the BC whereas the other involves the quasi-conformal map. Alternating minimization scheme can then be applied to minimize the energy functional. The proposed method significantly speeds up the previous BHF approach. It also extends the previous BHF algorithm to Riemann surfaces of arbitrary topologies, such as multiply-connected shapes. Experiments have been carried out on synthetic together with real surface data, which demonstrate the efficiency and efficacy of the proposed algorithm to solve the DOP.