Two-dimensional generalisations of dynamic programming for image analysis

  • C. A. GlasbeyEmail author


Dynamic programming (DP) is a fast, elegant method for solving many one-dimensional optimisation problems but, unfortunately, most problems in image analysis, such as restoration and warping, are two-dimensional. We consider three generalisations of DP. The first is iterated dynamic programming (IDP), where DP is used to recursively solve each of a sequence of one-dimensional problems in turn, to find a local optimum. A second algorithm is an empirical, stochastic optimiser, which is implemented by adding progressively less noise to IDP. The final approach replaces DP by a more computationally intensive Forward-Backward Gibbs Sampler, and uses a simulated annealing cooling schedule. Results are compared with existing pixel-by-pixel methods of iterated conditional modes (ICM) and simulated annealing in two applications: to restore a synthetic aperture radar (SAR) image, and to warp a pulsed-field electrophoresis gel into alignment with a reference image. We find that IDP and its stochastic variant outperform the remaining algorithms.


Forward-backward Gibbs sampler Image restoration Image warping Markov random field Pulsed-field gel electrophoresis Simulated annealing Synthetic aperture radar 


  1. Bellman, R.: Dynamic Programming. Princeton University Press, Princeton (1957) Google Scholar
  2. Besag, J.: On the statistical analysis of dirty pictures (with discussion). J. R. Stat. Soc. Ser. B 48, 259–302 (1986) zbMATHMathSciNetGoogle Scholar
  3. Besag, J., Kooperberg, C.: On conditional and intrinsic autoregressions. Biometrika 82, 733–746 (1995) zbMATHMathSciNetGoogle Scholar
  4. Boykov, Y., Veksler, O., Zabih, R.: Fast approximate energy minimization via graph cuts. IEEE Trans. Pattern Anal. Mach. Intell. 23, 1222–1239 (2001) CrossRefGoogle Scholar
  5. Eddy, S.R.: Multiple alignment using hidden Markov models. In: Rawlings, C., Clark, D., Altman, R., Hunter, L., Lengauer, T., Wodak, S. (eds.) Proceedings of the Third International Conference on Intelligent Systems for Molecular Biology, pp. 114–120. AAAI Press, Menlo Park (1995) Google Scholar
  6. Friel, N., Rue, H.: Recursive computing and simulation-free inference for general factorizable models. Biometrika 94, 661–672 (2007) zbMATHCrossRefMathSciNetGoogle Scholar
  7. Geman, S., Geman, D.: Stochastic relaxation, Gibbs distributions and the Bayesian restoration of images. IEEE Trans. Pattern Anal. Mach. Intell. 6, 721–735 (1984) zbMATHCrossRefGoogle Scholar
  8. Glasbey, C.A., Horgan, G.W.: Image Analysis for the Biological Sciences. Wiley, Chichester (1995) zbMATHGoogle Scholar
  9. Glasbey, C.A., Jones, R.: Fast computation of moving average and related filters in octagonal windows. Pattern Recogn. Lett. 18, 555–565 (1997) CrossRefGoogle Scholar
  10. Glasbey, C.A., Vali, L., Gustafsson, J.S.: A statistical model for unwarping of 1-D electrophoresis gels. Electrophoresis 26, 4237–4242 (2005) CrossRefGoogle Scholar
  11. Glasbey, C.A., Young, M.J.: Maximum a posteriori estimation of image boundaries by dynamic programming. Appl. Stat. 51, 209–221 (2002) zbMATHMathSciNetGoogle Scholar
  12. Greig, D.M., Porteous, B.T., Seheult, A.H.: Exact maximum a posteriori estimation for binary images. J. R. Stat. Soc. Ser. B 51, 271–279 (1989) Google Scholar
  13. Gustafsson, J.: Unwarping and analysing electrophoresis gels. PhD thesis, Department of Mathematical Sciences, Chalmers University of Technology, Gothenburg, Sweden (2005) Google Scholar
  14. Kirkpatrick, S., Gelatt, C.D., Vecchi, M.P.: Optimization by simulated annealing. Science 220, 671–680 (1983) CrossRefMathSciNetGoogle Scholar
  15. Künsch, H.R.: Robust priors for smoothing and image restoration. Ann. Inst. Stat. Math. 46, 1–19 (1994) zbMATHCrossRefGoogle Scholar
  16. Leung, C., Appleton, B., Sun, C.: Fast stereo matching by Iterated Dynamic Programming and quadtree subregioning. In: Hoppe, A., Barman, S., Ellis, T. (eds.) British Machine Vision Conference, vol. 1, pp. 97–106 (2004) Google Scholar
  17. MacDonald, I.L., Zucchini, W.: Hidden Markov and Other Models for Discrete-Valued Time Series. Chapman and Hall, London (1997) zbMATHGoogle Scholar
  18. Navajas, E.A., Glasbey, C.A., McLean, K.A., Fisher, A.V., Charteris, A.J.L., Lambe, N.R., Bunger, L., Simm, G.: In vivo measurements of muscle volume by automatic image analysis of spiral computed tomography scans. Anim. Sci. 82, 545–553 (2006) CrossRefGoogle Scholar
  19. Notredame, C.: Recent progress in multiple sequence alignment: a survey. Pharmacogenomics 3, 131–144 (2002) CrossRefGoogle Scholar
  20. Oliver, C.J.: Information from SAR images. J. Phys. D Appl. Phys. 24, 1493–1514 (1991) CrossRefGoogle Scholar
  21. Roberts, G.O., Sahu, S.K.: Updating schemes, correlation structure, blocking and parameterization for the Gibbs sampler. J. R. Stat. Soc. Ser. B 59, 291–317 (1997) zbMATHCrossRefMathSciNetGoogle Scholar
  22. Scott, S.L.: Bayesian methods for Hidden Markov Models: recursive computing in the 21st century. J. Am. Stat. Assoc. 97, 337–351 (2002) zbMATHCrossRefGoogle Scholar
  23. Storvik, G., Dahl, G.: Lagrangian-based methods for finding MAP solutions for MRF models. IEEE Trans. Image Process. 9, 469–479 (2000) CrossRefGoogle Scholar
  24. Szeliski, R., Zabih, R., Scharstein, D., Veksler, O., Kolmogorov, V., Agarwala, A., Tappen, M., Rother, C.: A comparative study of energy minimization methods for Markov random fields. In: Computer Vision—ECCV, Part 2. Lecture Notes in Computer Science, vol. 3952, pp. 16–29. Springer, Berlin (2006) Google Scholar

Copyright information

© Springer Science+Business Media, LLC 2008

Authors and Affiliations

  1. 1.Biomathematics and Statistics ScotlandEdinburghUK

Personalised recommendations