This Article 
   
 Share 
   
 Bibliographic References 
   
 Add to: 
 
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb
 
 Search 
   
Structural Matching in Computer Vision Using Probabilistic Relaxation
August 1995 (vol. 17 no. 8)
pp. 749-764

Abstract—In this paper, we develop the theory of probabilistic relaxation for matching features extracted from 2D images, derive as limiting cases the various heuristic formulae used by researchers in matching problems, and state the conditions under which they apply. We successfully apply our theory to the problem of matching and recognizing aerial road network images based on road network models and to the problem of edge matching in a stereo pair. For this purpose, each line network is represented by an attributed relational graph where each node is a straight line segment characterized by certain attributes and related with every other node via a set of binary relations.

[1] D.H. Ackley,G.E. Hinton,, and T.J. Sejnowski,“A learning algorithm for Bozmann machines,” Cognitive Science, vol. 9, pp. 147-169, 1985.
[2] H.S. Baird, Model-Based Image Matching Using Location. Cambridge, Mass.: MIT Press, 1985.
[3] D.H. Ballard and C.M. Brown, Computer Vision, Prentice Hall, Upper Saddle River, N.J., 1982.
[4] P.J. Besl and R. Jain,“Three-dimensional object recognition,” ACM Computing Surveys, vol. 17, pp. 75-145, Mar. 1985.
[5] J.R. Beveridge and E.M. Riseman,“Hybrid weak-perspective and full-perspective matching,” Conf. Computer Vision and Pattern Recognition, pp. 432-438, 1992.
[6] B. Bhanu,“Representation and shape matching of 3-D objects,” IEEE Trans.Pattern Analysis and Machine Intelligence, vol. 6, no. 3, pp. 340-351, 1984.
[7] B. Bhanu and O.D. Faugeras,“Shape matching of two-dimensional objects,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 6, pp. 137-156, 1984.
[8] E. Bienenstock,“Neural-like graph-matching techniques for image processing,” D.Z. Anderson, ed., Neural Information Processing Systems, pp. 211-235,Addison-Wesley, 1988.
[9] A. Blake,“, The least disturbance principle and weak constraints,” Pattern Recognition Letters, vol. 1, pp. 393-399, 1983.
[10] A. Blake and A. Zisserman, Visual Reconstruction. MIT Press, 1987.
[11] R.C. Bolles,“Robust feature matching through maximal cliques,” Proc. Soc. Photo-Optical Instrument Engineers, vol. 182, pp. 140-149, Apr. 1979.
[12] 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.
[13] T.A. Cass,“Polynomial-time object recognition in the presence of clutter, occlusion, and uncertainty,” Second European Conf. Computer Vision,Santa Margherita Ligure, Italy, May 1992, pp. 834-842. Springer-Verlag, 1992.
[14] L.S. Davis,“Shape matching using relaxation techniques,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 1, no. 1, pp. 60-72, Jan. 1979.
[15] O.D. Faugeras and M. Berthod,“Improving consistency and reducing ambiguity in stochastic labeling: An optimization approach,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 3, pp. 412-423, Apr. 1981.
[16] O.D. Faugeras and M. Hebert,“The representation, recognition, and locating of 3D objects,” Int’l J. of Robotics Research, vol. 5, no. 3, pp. 27-52, Fall 1986.
[17] M. Fischler and R. Elschlager,“The representation and matching of pictorial structures,” IEEE Trans. on Computers, vol. 22, pp. 67-92, 1973.
[18] 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.
[19] S. Geman and D. Geman,“Stochastic relaxation, Gibbs distributions, and the Bayesian restoration of images,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 6, pp. 721-741, 1984.
[20] D.E. Gharhaman,A.K.C. Wong,, and T. Au,“Graph optimal monomorphism algorithms,” IEEE Trans. Systems, Man, and Cybernetics, vol. 10, no. 4, pp. 181-188, Apr. 1980.
[21] B. Gidas, "A Renormalization Group Approach to Image Processing Problems," IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 11, pp. 164-180, 1989.
[22] W.E.L. Grimson and T. Lozano-Perez, “Localizing Overlapping Parts by Searching the Interpretation Tree,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 9, no. 4, pp. 469-482, Apr. 1987.
[23] E.R. Hancock and J. Kittler, "Edge-Labeling Using Dictionary-Based Relaxation," IEEE Trans Pattern Analysis and Machine Intelligence, vol. 12, no. 2, pp. 165-181, 1990.
[24] J.J. Hopfield,“Neurons with graded response have collective computational properties like those of two-state neurons,” Proc. National Academy of Science,USA, vol. 81, pp. 3,088-3,092, 1984.
[25] R.A. Hummel and S.W. Zucker,“On the foundations of relaxation labeling process,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 5, no. 3, pp. 267-286, May 1983.
[26] R.L. Kirby,“A product rule relaxation method,” CGIP, vol. 13, pp. 158-189, 1980.
[27] S. Kirkpatrick,C.D. Gellatt,, and M.P. Vecchi,“Optimisation by simulated annealing,” Science, vol. 220, pp. 671-680, 1983.
[28] J. Kittler and J. Foglein, "On Compatibility and Support Functions in Probabilistic Relaxation," Computer. Vision, Graphics, and Image Processing, 1986.
[29] J. Kittler and E.R. Hancock,“Combining evidence in probabilistic relaxation,” Int’l J. Pattern Recognition and Artificial Intelligence, vol. 3, pp. 29-51, 1989.
[30] C. Koch,J. Marroquin,, and A. Yuille,“Analog’neuronal’networks in early vision,” Proc. National Academy of Science,USA, vol. 83, pp. 4,263-4,267, 1986.
[31] S.Z. Li,“Matching: invariant to translations, rotations and scale changes,” Pattern Recognition, vol. 25, pp. 583-594, 1992.
[32] S.Z. Li,J. Kittler,, and M. Petrou,“On the automatic registration of aerial photographs and digitised maps,” Optical Engineering, vol. 32, no. 6, pp. 1,213-1,221, June 1993.
[33] S. Peleg,“A new probabilistic relaxation scheme,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 2, pp. 362-369, 1980.
[34] M. Petrou,“Optimal convolution filters and an algorithm for the detection of wide linear features,” IEEE Proc.-I, vol. 140, no. 5, pp. 331-339, Oct. 1993.
[35] B. Radig,“Image sequence analysis using relational structures,” Pattern Recognition, vol. 17, pp. 161-167, 1984.
[36] A. Rosenfeld,R. Hummel,, and S. Zucker,“Scene labeling by relaxation operations,” IEEE Trans. Systems, Man, and Cybernetics, vol. 6, pp. 420-433, June 1976.
[37] L.G. Shapiro and R.M. Haralick,“Structural description and inexact matching,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 3, pp. 504-519, Sept. 1981.
[38] F. Stein and G. Medioni, “Structural Indexing: Efficient 3-D Object Recognition,” IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 14, no. 2, pp. 125–145, 1992.
[39] S. Ullman,“Relaxation and constraint optimization by local process,” Computer Graphics and Image Processing, vol. 10, pp. 115-195, 1979.
[40] N.M. Vaidya and K.L. Boyer,“Stereopsis and image registration from extended range features in the absence of camera pose information,” Conf. Computer Vision and Pattern Recognition, pp. 76, 82.
[41] W.M. Wells,“MAP model matching,” Conf. Computer Vision and Pattern Recognition, pp. 486-492, 1991.
[42] A. Witkin,D. Terzopoulos,, and M. Kass,“Signal matching through scale space,” Int’l J. Computer Vision, vol. 1, pp. 133-144, 1987.
[43] K. Yamamoto,“A method of deriving compatibility coefficients for relaxation operators,” Computer Graphics and Image Processing, vol. 10, pp. 256-271, 1979.
[44] B. Yang, W.E. Snyder, and G.L. Bilbro, "Matching Oversegmented 3D Images to Models Using Associated Graphs," Image and Vision Computing, vol. 7, no. 2, pp. 135-143, 1989.
[45] S. Zucker, and J. Mohammed,“Analysis of probabilistic labeling process,” IEEE PRIP Conf.,Chicago, USA, pp. 167-173, 1978.

Index Terms:
Matching, probabilistic relaxation, object recognition.
Citation:
William J. Christmas, Josef Kittler, Maria Petrou, "Structural Matching in Computer Vision Using Probabilistic Relaxation," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 17, no. 8, pp. 749-764, Aug. 1995, doi:10.1109/34.400565
Usage of this product signifies your acceptance of the Terms of Use.