Abstract
We present and compare two new techniques for learning Relational Structures (RSs) as they occur in 2D pattern and 3D object recognition. These techniques, namely, Evidence-Based Networks (EBS-NNets) and Rulegraphs combine techniques from computer vision with those from machine learning and graph matching. The EBS-NNet has the ability to generalize pattern rules from training instances in terms of bounds on both unary (single part) and binary (part relation) numerical features. It also learns the compatibilities between unary and binary feature states in defining different pattern classes. Rulegraphs check this compatibility between unary and binary rules by combining evidence theory with graph theory. The two systems are tested and compared using a number of different pattern and object recognition problems.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
References
L. Shapiro and R. Haralick, “Structural descriptions and inexact matching,”IEEE Trans. Pattern Anal. Machine Intell. vol. 3, pp. 504–519, 1981.
J.R. Quinlan, “Induction of decision trees,”Machine Learning vol. 1, pp. 81–106, 1986.
J.R. Quinlan, “Learning logical definitions from relations,”Machine Learning vol. 5, pp. 239–266, 1990.
P. Flynn and A. Jain, “Three-dimensional object recognition,” inHandbook of Pattern Recognition and Image Processing, Volume 2: Computer Vision edited by Tzay Y. Young, Academic Press: New York, 1993.
A. Jain and D. Hoffman, “Evidence-based recognition of objects,”IEEE Trans. Pattern Anal. Machine Intell vol. 10, pp. 783–802, 1988.
T. Caelli and A. Pennington, “An improved rule generation method for evidence-based classification systems,”Pattern Recognition vol. 26, no. 5, pp. 733–740, 1993.
W.E.L. Grimson,Object Recognition by Computer MIT Press: Cambridge, MA, 1990.
K. Ikeuchi and T. Kanade, “Automatic generation of object recognition probrams,”Proc. IEEE vol. 76, no. 8, pp. 1016–1035, 1988.
R. Bolles and P. Horand, “3DPO: A three-dimensional part orientation system,”Int. J. Robot. Res. vol. 5, pp. 3–26, 1986.
M.A. Fischler and R.A. Elschlager, “The representation and matching of pictorial structures,”IEEE Trans. Comput. vol. 22, pp. 67–92, 1973.
R. Mohan and R. Nevatia, “Using perceptual organization to extract 3-D structures,”IEEE Trans. Pattern Anal. Machine Intell. vol. 11, pp. 1121–1139, 1989.
D.G. Lowe, “Three-dimensional object recognition from single two-dimensional images,”Artif. Intell. vol. 31, pp. 355–395, 1987.
R. Michalski and R.E. Stepp, “Automated construction of classifications: conceptual clustering versus numerical taxonomy,”IEEE Trans. Pattern Anal. Machine Intell. vol. 5, pp. 396–409, 1983.
N. Yokoya and M. Levine, “Range image segmentation based on differential geometry: a hybrid approach,”IEEE Trans. Pattern Anal. Machine Intell. vol. 11, pp. 643–649, 1989.
E. Aarts and J. Korst,Simulated Annealing and Boltzmann Machines Wiley: New York, 1989.
G. Hinton, “Connectionist learning procedures,”Artif. Intell. vol. 40, p. 1, 1989.
R.P. Lippmann, “An introduction to computing with neural nets,”IEEE:ASSP Mag., April, pp. 4–22, 1987.
A. Pearce, T. Caelli, and W. Bischof, “Rulegraphs for graph matching in pattern recognition,” Technical Report, Department of Computer Science, The University of Melbourne, Parkville 3053, Australia, 1993.
J. Hartigan,Clustering Algorithms Wiley: New York, 1975.
R. Tarjan and A. Trojanowski, “Finding the maximum independent set,”SIAM J. Comput. vol. 6, pp. 536–546, 1977.
T. Caelli and A. Dreier, “Variations on the evidence-based object recognition theme,”Pattern Recognition vol. 27, pp. 185–204, 1994.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Pearce, A.R., Caelli, T. & Bischof, W.F. Learning relational structures: Applications in computer vision. Appl Intell 4, 257–268 (1994). https://doi.org/10.1007/BF00872092
Issue Date:
DOI: https://doi.org/10.1007/BF00872092