This Article 
   
 Share 
   
 Bibliographic References 
   
 Add to: 
 
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb
 
 Search 
   
A Renormalization Group Approach to Image Processing Problems
February 1989 (vol. 11 no. 2)
pp. 164-180

A method for studying problems in digital image processing, based on a combination of renormalization group ideas, the Markov random-field modeling of images, and metropolis-type Monte Carlo algorithms, is presented. The method is efficiently implementable on parallel architectures, and provides a unifying procedure for performing a hierarchical, multiscale, coarse-to-fine analysis of image-processing tasks such as restoration, texture analysis, coding, motion analysis, etc. The method is formulated and applied to the restoration of degraded images. The restoration algorithm is a global-optimization algorithm applicable to other optimization problems. It generates iteratively a multilevel cascode of restored images corresponding to different levels of resolution, or scale. In the lower levels of the cascade appear the large-scale features of the image, and in the higher levels, the microscopic features of the image.

[1] J. Besag, "Efficiency of pseudo-likelihood estimations for simple Gaussian fields,"Biometrika, vol. 64, pp. 616-618, 1977.
[2] O. J. Braddock, F. W. Campbell, and J. Atkinson, "Channels in vision: Basic aspects," inHandbook of Sensory Physiology: Perception, vol. 8, R. Held, H. W. Leibowitz, and H. L. Teuber, Eds.. 1978.
[3] P. J. Burt, "Fast filter transformers for image processing,"Comput. Graphics Image Processing, vol. 16, pp. 20-51, 1981.
[4] P. J. Burt, "The pyramid as a structure for efficient computation," inMultiresolution Image Processing and Analysis, A. Rosenfeld, Eds. New York: Springer-Verlag, 1984.
[5] P. J. Burt, and G. Sperling, "Time, distance, and feature trade-offs in visual apparent motion,"Psychol. Rev., vol. 88, pp. 171-195, 1981.
[6] R. Chellappa and R. L. Kashyap, "Digital image restoration using spatial interaction models,"IEEE Trans. Acoust., Speech, Signal Processing, vol. ASSP-30, pp. 461-472, 1982.
[7] D. B. Cooper and F. D. Sung, "Multi-window parallel adaptive boundary finding in computer vision,"IEEE Trans. Pattern Anal., Machine Intell., vol. PAMI-5, pp. 299-316, 1983.
[8] A. P. Dempster, N. Laird, and D. Rubin, "Maximum likelihood from incomplele data via the EM algorithm,"J. Roy Stat. Soc B, vol. 39, pp. 1-38, 1977.
[9] H. Derin, H. Elliott, R. Cristi, and D. Geman, "Bayes smoothing algorithms for segmentation of binary images modeled by Markov fields,"IEEE Trans. Pattern Anal. Machine Intell., vol. PAMI-6, pp. 707-720, 1984.
[10] S. Gelfand and S. K. Mitter, "Analysis of simulated annealing for optimization," inProc. IEEE Conf. Decision and Control., 1985, pp. 779-786; also M.I.T. Tech. Rep. LIDS-1494, 1985.
[11] M. Gell-Mann and F. E. Low, "Quantum electrodynamics at small distances,"Phys. Rev., vol. 95, pp. 1300-1312, 1954.
[12] S. Geman and D. Geman, "Stochastic relaxation, Gibbs distributions, and the Bayesian restoration of images,"IEEE Trans. Pattern Anal. Machine Intel., vok PAMI-6, pp. 721-741, 1984.
[13] S. Geman and C. Graffigne, "Markov random field image models and their applications to computer vision," inProc. 1986 Int. Congr. Mathematicians, A. M. Gleason, Ed., AMS, New York, 1987.
[14] S. Geman and D. E. McClure, "Bayesian image analysis: An application to single photon emission tomography," in1985 Proc. Amer. Stat. Ass., Stat. Comput. Section.
[15] B. Gidas, "Non-stationary Markov chains and convergence of the annealing algorithm,"J. Stat. Phys., vol. 39, pp. 73-131, 1985.
[16] B. Gidas, "Consistency of maximum likelihood and pseudo-likelihood estimators for Gibbs distributions," inProc. Workshop Stochastic Differential Systems with Applications, IMA, Univ. Minnesota. New York: Springer-Verlag, 1987.
[17] B. Gidas, "Efficient methods for computing coarse-grid clique parameters for the renormalization group algorithm," Div. Appl. Math., Brown Univ., Tech. Rep., 1987.
[18] J. Glimm and A. Jaffe,Quantum Physics. New York: Springer-Verlag, 1981.
[19] U. Grenander,Tutorial in Pattern Theory, Brown Univ., 1983.
[20] A. Habibi, "Two-dimensional Bayesian estimate of images,"Proc. IEEE, vol. 60, pp. 878-883, 1972.
[21] W. Hackbush,Multi-Grid Methods and Applications. New York: Springer-Verlag, to be published.
[22] B. Hajek, "Cooling schedules for optimal annealing,"Math. Oper. Res., vol. 13, no. 2, pp. 311-329, May 1988.
[23] A. R. Hansen and H. Elliott, "Image segmentation using simple Markov field models,"Comput. Graphics Image Processing, vol. 20, pp. 101-132, 1982.
[24] A. R. Hanson and E. M. Roseman, "Segmentation of natural scenes," inComput Vision Systems. New York: Academic, 1978.
[25] B. R. Hunt, "Bayesian methods in nonlinear digital image restoration,"IEEE Trans. Comput., vol. C-23, pp. 219-229, 1977.
[26] G. E. Hinlon, T. J. Sejnowski, and D. H. Ackley, "Boltzmann machines: Constraint satisfaction networks and learns," Carnegie-Mellon Tech. Rep. CMU-CS-84-119, 1984.
[27] A. K. Jain, "Advances in mathematical models for image processing,"Proc. IEEE, vol 69, pp. 502-528, 1981.
[28] A. K. Jain and E. Angel, "Image restoration, modeling and reduction of dimensionality,"IEEE Trans. Comput., vol. C-23, pp. 470-476, 1974.
[29] L. P. Kadanoff, "Notes on Migdal's recursion formulas,"Ann. Phys., vol. 100, pp. 359-394, 1976.
[30] L. P. Kadanoff and A. Houghton, "Numerical evaluations of the critical properties of the two-dimensional Ising model,"Phys. Rev., vol. B11, pp. 377-386, 1975.
[31] R. Kashyap and R. Chellapa, "Estimation and choice of neighbors in spatial interaction models of images,"IEEE Trans. Inform. Theory, vol. IT-29, pp. 60-72, Jan. 1983.
[32] S. Kirpatrick, C. D. Gelatt, and M. Vecchi, "Optimization of sim ulated annealing,"Science, vol. 220, pp. 621-680, May 1983.
[33] N. Metroplis, A. W. Rosenbluth, M. N. Rosenbluth, A. H. Teller, and E. Teller, "Equations of state calculations by fast computing machines,"J. Chem. Phys., vol. 21, pp. 1087-1091, 1953.
[34] A. A. Migdal, "'Phase transitions in gauge and spin-lattice systems,"Sov. Phys. JETP, vol. 42, pp. 743-746, 1976.
[35] D. Mitra, F. Romeo, and A. Sangiovanni-Vincentelli, "Convergence and finite-time behavior of simulated annealing," inProc. 24th Conf. Decision and Control 1985, pp. 761-767.
[36] K. Binder, Ed.,Monte-Carlo Methods in Statistical Physics. New York: Springer-Verlag, 1979.
[37] A. Rosenfeld, Ed.,Multiresolution Image Processing and Analysis. New York: Springer-Verlag, 1984.
[38] N. E. Nahi, and T. Assefi, "Bayesian recursive image estimation,"IEEE Trans. Comput., vol. C-21, pp. 734-738, 1972.
[39] T. Niemeyer and J. M. J. Van Leeuwen, "Wilson's theory for 2-dimensional Ising spin systems,"Physics, vol. 71, pp. 17-40, 1974.
[40] A. Rosenfeld and A. Kak,Digital Picture Processing, New York: Academic, 1976.
[41] E. Stueckelberg and A. Petermann, "La normalization des constantes dans la theorie des quanta,"Helv. Phys. Acta, vol. 26, pp. 489-520, 1953.
[42] D. Terzopoulos, "Multilevel computational processes for visual surface reconstruction,"Comput. Vision, Graphics, Image Processing, vol. 24, pp. 52-56, 1983.
[43] H. J. Trussel, "The relationship between image restoration by the maximum a posterior method and a maximum entropy method,"IEEE Trans. Acoust., Speech, Signal Processing, vol. ASSP-28, pp. 114- 117, 1980.
[44] C. W. Tyler, "Stereoscopic vision: Cortical limitations and a disparity scaling effect,"Science, vol. 181, pp. 276-278, 1973.
[45] C. W. Tyler and J. J. Chang, "Visual echoes: The perception of repetition in quasi-random patterns,"Vision Res., vol. 17, pp. 109-116, 1977.
[46] J. M. J. van Leeuwen, "Singularities in the critical surface and universality for Ising-like spin systems,"Phys. Rev. Lett., vol. 34, pp. 1056-1058, 1975.
[47] K. Wilson, "The renormalization group: Critical phenomena and the Kondo problem,"Rev. Mod. Phys., vol. 47, pp. 773-840, 1975.
[48] K. Wilson and J. Kogut, "The renormalization group and theε-expansion,"Phys. Rep., vol. C12, pp. 75-200, 1974.
[49] A. P. Witkin, "Scale space filtering: A new approach to multi-scale description," inImage Understanding 1984, S. Ullman and W. Richards, Eds. Norwood, NJ: Ablex, 1984.
[50] A. L. Yuille and T. Poggio, "Scaling theorems for zero-crossings," M.I.T. Artificial Intell. Lab., Memo 772, June 1983.

Index Terms:
image restoration; computerised picture processing; renormalization group; digital image processing; Markov random-field modeling; Monte Carlo algorithms; coarse-to-fine analysis; texture analysis; global-optimization; multilevel cascode; computerised picture processing; Markov processes; Monte Carlo methods; optimisation
Citation:
B. Gidas, "A Renormalization Group Approach to Image Processing Problems," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 11, no. 2, pp. 164-180, Feb. 1989, doi:10.1109/34.16712
Usage of this product signifies your acceptance of the Terms of Use.