Abstract
The problem to recognize objects that form an array of interrelated data is investigated. In the problem of machine learning, the array components belong to some class of a finite set. In this paper the interrelationship of array elements is presented by its adjacency graph. An efficient noniterative recognition algorithm for restoring an a posteriori marginal distributions of hidden classes for array elements is developed for a treelike adjacency graph. This algorithm modifies for each array element the hidden class distribution obtained as a result of learning for independent objects. Usually arbitrary graphs for real data contain cycles, for example, the rectangular adjacency lattice of points for 2D raster images or 3D seismic data. The treelike approximation of such graphs inevitably strongly distorts the interrelations between array elements. In the present paper, the reduced set of interrelations between array elements is balanced by an extended set of acyclic graphs themselves. By the example of the segmentation problem for texture raster images, we investigate a set of acyclic graphs and present the experimental results.
Similar content being viewed by others
References
L. R. Rabiner, “A Tutorial on Hidden Markov Models and Selected Applications in Speech Recognition,” Proc. IEEE 77(2), 257–286 (1977).
S. Geman and D. Geman, “Stochastic Relaxation, Gibbs Distributions, and the Bayesian Restoration of Images,” IEEE Trans. PAMI 6 721–741 (1984).
M. I. Schlesinger and B. Flach, “Some Solvable Subclasses of Structural Recognition Problems,” in Proc. Czech Pattern Recognition Workshop (Prague, 2000), pp. 55–62.
S. Z. Li, Markov Random Field Modeling in Image Analysis (Springer-Verlag, London, 2009).
V. V. Mottl, S. D. Dvoenko, V. B. Levyant, and I. B. Muchnik, “Pattern Recognition in Spatial Data: A New Method of Seismic Explorations for Oil and Gas in Crystalline Basement Rocks,” in Proc. 15th ICPR’2000 (Barcelona, 2000), Vol. 3, pp. 210–213.
V. V. Mottl, S. D. Dvoenko, and A. V. Kopylov, “Pattern Recognition in Interrelated Data: The Problem, Fundamental Assumptions, Recognition Algorithms,” in Proc. 17th ICPR (Cambridge, 2004), Vol. 1, pp. 188–191.
S. D. Dvoenko, A. V. Kopylov, and V. V. Mottl, “The Problem of Pattern Recognition in Arrays of Interconnected Objects. Statement of the Recognition Problem and Basic Assumptions,” Automat. Remote Control 65(1), 127–141 (2004).
S. D. Dvoenko, A. V. Kopylov, and V. V. Mottl, “A Problem of Pattern Recognition in Arrays of Interrelated Objects. Recognition Algorithm,” Automat. Remote Control 66(12), 2019–2032 (2005).
V. Kolmogorov, “Convergent Tree-Reweighed Message Passing for Energy Minimization,” IEEE Trans. PAMI 10, 1568–1583 (2006).
R. Szeliski, R. Zabih, D. Scharstein, O. Veksler, V. Kol- mogorov, A. Agarwala, M. Tappen, and C. Rother, “A Comparative Study of Energy Minimization Methods for Markov Random Fields with Smoothness-Based Priors,” IEEE Trans. PAMI 6, 1068–1080 (2007).
M. J. Wainwright and M. I. Jordan, “Graphical Models, Exponential Families, and Variational Inference,” Foundations Trends Mach. Learn. 1, 1–305 (2008).
J. G. Kemeny and J. L. Snell, Finite Markov Chains (Springer-Verlag, New York, 1983).
W. Feller, An Introduction to Probability Theory and Its Applications (Wiley, New York, 1970), Vol. 1.
Author information
Authors and Affiliations
Corresponding author
Additional information
Sergey Danilovich Dvoenko. Graduated from the postgraduate courses of the Institute of Control Science, Russian Academy of Sciences. Received candidate’s degree in 1992. He received the associated professor designation in 1998. Graduated from the doctoral courses of Tula State University and received doctor’s degree in 2002 at the Dorodnitsyn Computing Centre, Russian Academy of Sciences. Professor at Tula State University (Automation and Remote Control Department). Member of the Russian organization “Association for Pattern Recognition and Image Analysis.” His scientific interests include the following fields: machine learning and pattern recognition, cluster-analysis and data mining, image processing, hidden Markov models and fields in applied problems.
Rights and permissions
About this article
Cite this article
Dvoenko, S.D. Recognition of dependent objects based on acyclic Markov models. Pattern Recognit. Image Anal. 22, 28–38 (2012). https://doi.org/10.1134/S1054661812010130
Received:
Published:
Issue Date:
DOI: https://doi.org/10.1134/S1054661812010130