This Article 
   
 Share 
   
 Bibliographic References 
   
 Add to: 
 
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb
 
 Search 
   
Structural Matching by Discrete Relaxation
June 1997 (vol. 19 no. 6)
pp. 634-648

Abstract—This paper describes a Bayesian framework for performing relational graph matching by discrete relaxation. Our basic aim is to draw on this framework to provide a comparative evaluation of a number of contrasting approaches to relational matching. Broadly speaking there are two main aspects to this study. Firstly we focus on the issue of how relational inexactness may be quantified. We illustrate that several popular relational distance measures can be recovered as specific limiting cases of the Bayesian consistency measure. The second aspect of our comparison concerns the way in which structural inexactness is controlled. We investigate three different realizations of the matching process which draw on contrasting control models. The main conclusion of our study is that the active process of graph-editing outperforms the alternatives in terms of its ability to effectively control a large population of contaminating clutter.

[1] H.G. Barrow and R.J. Popplestone, "Relational Descriptions in Picture Processing," Machine Intelligence, vol. 6, pp. 377-396, 1971.
[2] H.G. Barrow and R.M Burstall, "Subgraph Isomorphism, Matching Relational Structures and Maximal Cliques," Information Processing Letters, vol. 4, pp. 83-84, 1976.
[3] K.L. Boyer and A.C. Kak,“Structural stereopsis in 3D vision,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 10, no. 2, pp. 144-166, 1988.
[4] W. Christmas, J. Kittler, and M. Petrou, "Probabilistic Feature-Labeling Schemes—Modeling Compatibility Coefficient Distribution," Image and Vision Computing, vol. 14, pp. 617-625, 1996.
[5] W.J. Christmas, J. Kittler, and M. Petrou, “Structural Matching in Computer Vision Using Probabilistic Relaxation,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 17, no. 8, pp. 749–764, Aug. 1995.
[6] S.J. Dickinson, A.P. Pentland, and A. Rosenfeld, "3D Shape Recovery Using Distributed Aspect Matching," IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 14, no. 2, pp. 174-198, Feb. 1992.
[7] O.D. Faugeras and K.E. Price, "Semantic Labeling of Aerial Images Using Stochastic Relaxation," IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 3, pp. 633-642, 1981.
[8] A.M. Finch, R.C. Wilson, and E.R. Hancock, "Matching Deluanay Graphs," Pattern Recognition, vol. 30, pp. 123-140, 1997.
[9] D. Fogel, "An Introduction to Simulated Evolutionary Optimization," IEEE Trans. Neural Networks, vol. 5, pp. 3-14, Jan. 1994.
[10] P.J. Flynn and A.K. Jain,“CAD-based computer vision: From CAD models to relational graphs,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 13, no. 2, pp. 114-132, Feb. 1991.
[11] E. Gardner, "Structure and Metastable States in the Hopfield Model," J. Physics, vol. A19, pp. L1,047-L1,052, 1986.
[12] D. Geman and S. Geman, "Stochastic Relaxation, Gibbs Distributions and Bayesian Restoration of Images," IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 6, pp. 721-741, 1984.
[13] D. Geiger and F. Girosi,“Parallel and deterministic algorithms from MRFs: Surface reconstruction,” IEEE Transactions on PAMI, vol. 13, no. 5, pp. 401-412, May 1991.
[14] F. Glover, "Genetic Algorithms and Tabu Search—Hybrids for Optimization," Discrete Applied Mathematics, vol. 49, pp. 111-134, 1995.
[15] F. Glover, "Tabu Search for Nonlinear and Parametric Optimisation (With Links to Genetic Algorithms)," Discrete Applied Mathematics, vol. 49, pp. 231-255, 1995.
[16] S. Gold and A. Rangarajan, “A Graduated Assignment Algorithm for Graph Matching,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 18, no. 4, pp. 377-388, Apr. 1996.
[17] E.R. Hancock and J. Kittler, "Discrete Relaxation," Pattern Recognition, vol. 23, pp. 711-733, 1990.
[18] E.R. Hancock and J. Kittler, "A Bayesian Interpretation for the Hopfield Network," 1993 IEEE Int'l Conf. Neural Networks, pp. 341-346, 1993.
[19] E.R. Hancock, "Resolving Edge-Line Ambiguities Using Probabilistic Relaxation," Proc. IEEE Computer Society Conf. Computer Vision and Pattern Recognition, pp. 300-306, 1993.
[20] R. Horaud and T. Skordas, “Stereo Correspondence Through Feature Grouping and Maximal Cliques,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 11, pp. 1,168-1,180, 1989.
[21] W. Kim and A.C. Kak, "3-D Object Recognition Using Bi-Partite Graph Matching Embedded in Discrete Relaxation," IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 13, pp. 224-251, 1991.
[22] B.T. Messmer and H. Bunke, "Efficient Error-Tolerant Subgraph Isomorphism Detection," D. Dori and A. Bruckstein, eds., Shape, Structure and Pattern Recognition, pp. 231-240, 1994.
[23] A. Rosenfeld, R. Hummel, and S. Zucker, "Scene Labeling by Relaxation Operations," IEEE SMC, vol. 6, pp. 420-433, 1976.
[24] A. Sanfeliu and K.S. Fu, "A Distance Measure Between Attributed Relational Graph," IEEE SMC, vol. 13, pp. 353-362, 1983.
[25] K. Sengupta and K.L. Boyer, “Organizing Large Structural Modelbases,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 17, no. 4, pp. 321-332, Apr. 1995.
[26] L. Shapiro and R.M. Haralick, "Structural Description and Inexact Matching," IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 3, pp. 504-519, 1981.
[27] L. Shapiro and R.M. Haralick, "A Metric for Comparing Relational Descriptions," IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 7, pp. 90-94, 1985.
[28] P. Simic, "Constrained Nets for Graph Matching and Other Quadratic Assignment Problems," Neural Computation vol. 3, pp. 268-291, 1991.
[29] P.N. Suganathan, E.K. Teoh, and D.P. Mital, "Pattern Recognition by Graph Matching Using Potts MFT Networks," Pattern Recognition, vol. 28, pp. 997-1,009, 1995.
[30] M.L. Williams, R.C. Wilson, and E.R. Hancock, "Deterministic Search Strategies for Relational Graph Matching," Advances in Energy Minimisation Methods for Computer Vision, Lecture Notes in Computer Science, to appear.
[31] R.C. Wilson, A.N. Evans, and E.R Hancock, "Relational Matching by Discrete Relaxation," Image and Vision Computing, vol. 13, pp. 411-422, 1994.
[32] R.C. Wilson and E.R. Hancock, "Relational Matching With Dynamic Graph Structures" Proc. Fifth Int'l Conf. Computer Vision, pp. 450-456, 1995.
[33] R.C. Wilson and E.R Hancock, "A Bayesian Compatibility Model for Graph Matching," Pattern Recognition Letters, vol. 17, pp. 263-276, 1996.
[34] A.K.C. Wong and M. You, "Entropy and Distance of Random Graphs With Application to Structural Pattern Recognition," IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 7, pp. 509-609, 1985.
[35] D. Yang and J. Kittler, "MFT Based Discrete Relaxation for Matching High Order Relational Structures," Proc. 12th Int'l Conf. Pattern Recognition, vol. 2, pp. 219-223, 1994.

Index Terms:
Structural graph matching, discrete relaxation, energy minimization, Bayesian, graph edit, clutter, MAP estimation, SAR images, infrared images.
Citation:
Richard C. Wilson, Edwin R. Hancock, "Structural Matching by Discrete Relaxation," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 19, no. 6, pp. 634-648, June 1997, doi:10.1109/34.601251
Usage of this product signifies your acceptance of the Terms of Use.