Abstract
Recently, the problem of zoning design has been considered as an optimization problem and the optimal zoning is found as the one which minimizes the value of the cost function associated to the classification. For the purpose, well-suited zoning representation techniques based on Voronoi Diagrams have been proposed and effective real-coded genetic algorithms have been used for optimization.
In this paper, starts from the consideration that whatever zoning method is considered, the role of feature membership function is crucial, since it determines the influence of a feature to each zone of the zoning method. Thus, in the paper the role of feature membership functions in Voronoi-based zoning methods is investigated. For the purpose, abstract-level, ranked-level and measurement-level membership functions are considered and their effectiveness is estimated under different Voronoi-based zoning methods.
The experimental tests, carried out in the field of hand-written numeral recognition, show that the best results are obtained when specific measurement-level membership functions are used.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Trier, O.D., Jain, A.K., Taxt, T.: Feature Extraction Methods For Character Recognition – A Survey. Pattern Recognition 29(4), 641–662 (1996)
Plamondon, R., Srihari, S.N.: On-line and Off-line Handwriting Recognition: A comprehensive survey. IEEE Trans. Pattern Analysis and Machine Intelligence 22(1), 63–84 (2000)
Suen, C.Y., Guo, J., Li, Z.C.: Analysis and Recognition of Alphanumeric Handprints by Parts. IEEE Trans. Systems, Man and Cybernetics 24(4), 614–630 (1994)
Dimauro, G., Impedovo, S., Modugno, R., Pirlo, G.: Numeral recognition by weighting local decisions. In: Proc. ICDAR 2003, Edinburgh, UK, pp. 1070–1074 (August 2003)
Baptista, G., Kulkarni, K.M.: A high accuracy algorithm for recognition of hand-written numerals. Pattern Recognition 4, 287–291 (1988)
Lucchese, M.G., Impedovo, S., Pirlo, G.: Optimal Zoning Design by Genetic Algorithms. IEEE Trans. Sys. Men and Cybern. - Part A 36(5), 833–846 (2006)
Kuncheva, L.I., Jain, L.C.: Designing Classifier Fusion Systems by Genetic Algorithms. IEEE Trans. Evolut. Comput. 4(4) (2000)
Aurenhammer, F.: Voronoi Diagrams: A Survey of a Fundamental Geometric Data Structure. ACM Computing Surveys 3(3), 345–405 (1991)
de Berg, M., Schwarzkopf, O., van Kreveld, M., Overmars, M.: Computational Geometry: Algorithms and Applications. Springer, Berlin (2000)
Baeck, T.: Evolutionary Algorithms in Theory and Practice: Evolution Strategies, Evolution Programming, Genetic Algorithms. Oxford Univ. Press, New York (1996)
Michalewicz, Z.: Genetic Algorithms + Data Structure=Evolution Programs. Springer, Berlin (1996)
Grefenstette, J.J.: Optimization of control parameters for genetic algorithms. IEEE Trans. SMC 16(1), 122–128 (1986)
Back, T., Fogel, D., Michalewicz, Z. (eds.): Handbook of Evolutionary Computation. Institute of Physics Publishing Ltd., Bristol and Oxford University Press, New York (1997)
Hull, J.: A database for handwritten text recognition research. IEEE Trans. PAMI 16(5), 550–554 (1994)
Heutte, L., Paquet, T., Moreau, J.V., Lecourtier, Y., Olivier, C.: A Structural / Statistical Features Based Vector for Handwritten Character Recognition. Pattern Recognition Letters 9, 629–641 (1998)
Naccache, N.J., Shinghal, R.: SPTA: A proposed algorithm for thinning binary patterns. IEEE Trans. SMC 14(3), 409–418 (1994)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2009 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Impedovo, S., Ferrante, A., Modugno, R., Pirlo, G. (2009). Feature Membership Functions in Voronoi-Based Zoning. In: Serra, R., Cucchiara, R. (eds) AI*IA 2009: Emergent Perspectives in Artificial Intelligence. AI*IA 2009. Lecture Notes in Computer Science(), vol 5883. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-10291-2_21
Download citation
DOI: https://doi.org/10.1007/978-3-642-10291-2_21
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-10290-5
Online ISBN: 978-3-642-10291-2
eBook Packages: Computer ScienceComputer Science (R0)