This Article 
   
 Share 
   
 Bibliographic References 
   
 Add to: 
 
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb
 
 Search 
   
An Algebraic Approach to Automatic Construction of Structural Models
December 1993 (vol. 15 no. 12)
pp. 1298-1311

We present algebraic approach to the inductive learning of structural models and automatic construction of shape prototypes for character recognition on the basis of the algebraic description of curve structure proposed by Nishida and Mori (1991, 1992). A class in the structural models is a set of shapes that can be transformed continuously to each other. We consider an algebraic representation of continuous transformation of components of the shape, and give specific properties satisfied by each component in the class. The generalization rules in the inductive learning are specified from the viewpoints of continuous transformation of components and relational structure among the components. The learning procedure generalizes a pair of classes into one class incrementally and hierarchically in terms of the generalization rules. We show experimental results on handwritten numerals.

[1] H. S. Baird, "Feature extraction for hybrid structural/statistical pattern classification,"Comput. Vision Graphics Image Processing, vol. 42, pp. 318-333, 1988.
[2] H. S. Baird, H. Bunke, and K. Yamamoto, Eds.,Structured Image Analysis. New York: Springer Verlag, 1992.
[3] H. S. Baird and R. Fossey, "A 100-font classifier," inProc. 1st Int. Conf. Document Anal. Recognition, Saint Malo, France, Sept. 30-Oct. 2, 1991, pp. 332-340.
[4] T. G. Dietterich and R. S. Michalski, "Inductive learning of structural descriptions,"Artificial Intell., vol. 16, pp. 257-294, 1981.
[5] T. G. Dietterich and R. S. Michalski, "A comparative review of selected methods for learning from examples," inMachine Learning, vol. I, R. S. Michalski, J. G. Carbonell, and T. M. Mitchell, Eds. Los Altos, CA: Morgan Kaufmann, 1983, pp. 41-81.
[6] R. O. Duda and P. E. Hart,Pattern Classification and Scene Analysis. New York: Wiley, 1973.
[7] R. M. Haralick and G. L. Elliott, "Increasing tree search efficiency for constraint satisfaction problems,"Artificial Intell., vol. 14, pp. 263-313, 1980.
[8] R. M. Haralick and L. G. Shapiro, "The consistent labeling problem: part 1,"IEEE Trans. Pattern Anal. Mach. Intell., vol. PAMI-1, pp. 173-184, 1979.
[9] R. Kasturi and L. O'Gorman, Eds.,Machine Vision Applications, vol. 5, no. 3 (special issue on Document Image Analysis Techniques), Summer 1992.
[10] C. L. Liu,Elements of Discrete Mathematics. New York: McGraw-Hill, 1977.
[11] R. S. Michalski, "A theory and methodology of inductive learning," inMachine Learning, vol. I, R. S. Michalski, J. G. Carbonell, and T. M. Mitchell, Eds. Los Altos, CA: Morgan Kaufmann, 1983, pp. 41-81.
[12] C. Nash and S. Sen,Topology and Geometry for Physicists. London: Academic, 1983.
[13] H. Nishida, "A structural approach to character recognition," Doctoral dissertation, Dept. Math. Eng. Inform. Phys., Univ. of Tokyo, Tokyo, Japan, 1992.
[14] H. Nishida and S. Mori, "An approach to automatic construction of structural models for character-recognition," inProc. 1st Int. Conf. Document Anal. Recognition, Saint Malo, France, Sept. 30-Oct. 2, 1991, pp. 231-241.
[15] H. Nishida and S. Mori, "Algebraic description of curve structure,"IEEE Trans. Pattern Anal. Mach. Intell., vol. 14, no. 5, pp. 516-533, 1992.
[16] H. Nishida and S. Mori, "Structural analysis and description of curves by quasi-topological features and singular points," inStructured Document Image Analysis, H. S. Baird, H. Bunke, and K. Yamamoto, Eds. Heidelberg: Springer-Verlag, 1992, pp. 139-187.
[17] H. Nishida, T. Suzuki, and S. Mori, "Thin line representation from contour representation of handprinted characters," inFrom Pixels to Features III: Frontiers in Handwriting Recognition, S. Impedovo and J. C. Simon, Eds. Amsterdam: Elsevier, 1992, pp. 29-40.
[18] L. O'Gorman and R. Kasturi (Eds.),Computer, vol. 25, no. 7 (special issue on Document Image Analysis Systems), July 1992.
[19] T. Pavlidis,Structural Pattern Recognition. New York: Springer-Verlag, 1977.
[20] T. Pavlidis, personal communication, 1990.
[21] T. Pavlidis and S. Mori, Eds.,Proc. IEEE(Special Issue Optical Character Recognition), vol. 80, no. 7, 1992.
[22] J. R. Quinlan, "The effect of noise on concept learning," inMachine Learning, vol. II, R. S. Michalski, J. G. Carbonell, and T. M. Mitchell, Eds. Los Altos, CA: Morgan Kaufmann, 1986, pp. 149-166.
[23] S. Schroeder and L. G. Shapiro, "Report of working group on automatic construction of models," inStructural Pattern Analysis, R. Mohr, T. Pavlidis, and A. Sanfeliu, Eds. Singapore: World Scientific, 1990, pp. 255-260.
[24] L. G. Shapiro and R. M. Haralick, "Structural descriptions and inexact matching,"IEEE Trans. Pattern Anal. Mach. Intell., vol. PAMI-3, no. 5, pp. 504-519, 1981.
[25] T. Suzuki and S. Mori, "Structural description of line images by cross section sequence graph and its application to skeletonization,"Int. J. Pattern Recognition Artificial Intell., vol. 7, no. 5, 1993.
[26] P. H. Winston, "Learning structural descriptions from examples," inThe Psychology of Computer Vision, P. H. Winston, Ed. New York: McGraw-Hill, 1975, ch. 5.
[27] K. Yamamoto and S. Mori, "Recognition of handprinted characters by outermost point method,"Pattern Recognition, vol. 12, pp. 229-236, 1980.

Index Terms:
algebraic approach; structural model automatic construction; inductive learning; shape prototypes; character recognition; curve structure; algebraic representation; continuous transformation; handwritten numerals; algebra; learning (artificial intelligence); optical character recognition
Citation:
H. Nishida, S. Mori, "An Algebraic Approach to Automatic Construction of Structural Models," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 15, no. 12, pp. 1298-1311, Dec. 1993, doi:10.1109/34.250847
Usage of this product signifies your acceptance of the Terms of Use.