Fast Handwritten Recognition Using Continuous Distance Transformation

  • Joaquim Arlandis
  • Juan-Carlos Perez-Cortes
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2905)

Abstract

The Continuous Distance Transformation (CDT) used in conjunction with a k-NN classifier has been shown to provide good results in the task of handwriting recognition [1]. Unfortunately, efficient techniques such as kd-tree search methods cannot be directly used in the case of certain dissimilarity measures like the CDT-based distance functions. In order to avoid exhaustive search, a simple methodology which combines kd-trees for fast search and Continuous Distance Transformation for fine classification, is presented. The experimental results obtained show that the recognition rates achieved have no significant differences with those found using an exhaustive CDT-based classification, with a very important temporal cost reduction.

Keywords

Exhaustive Search Dissimilarity Measure White Pixel Fast Search Distance Transformation 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

References

  1. 1.
    Arlandis, J., Perez-Cortes, J.C., Llobet, R.: Handwritten Character Recognition Using Continuous Distance Transformation. In: Proceedings of the 15th. International Conference on Pattern Recognition, vol. 1, pp. 940–943 (2000)Google Scholar
  2. 2.
    Arlandis, J., Perez-Cortes, J.C.: The Continuous Distance Transformation: A Generalization of the Distance Transformation for Continuous-valued Images. Pattern Recognition and Applications 56, 89–98 (2000)Google Scholar
  3. 3.
    Arya, S., Mount, D.M., Netanyahu, N.S., Silverman, R., Wu, A.: An optimal algorithm for approximate nearest neighbor searching. Journal of the ACM 45, 891–923 (1998)MATHCrossRefMathSciNetGoogle Scholar
  4. 4.
    Beis, J.S., Lowe, D.G.: Indexing without Invariants in 3D Object Recognition. IEEE Trans. on PAMI 21(10), 1000–1015 (1999)Google Scholar
  5. 5.
    Bentley, J.L., Weide, B.W., Yao, A.C.: Optimal Expected Time Algorithms for Closest Point Problems. CM Trans. Math. Software 6, 563–580Google Scholar
  6. 6.
    Borgefors, C.: A New Distance Transformation Approximating the Euclidean Distance. In: Proceedings of the 8th. International Conference on Pattern Recognition, vol. 1, pp. 336–338 (1986)Google Scholar
  7. 7.
    Friedman, J.H., Bentley, J.L., Finkel, R.A.: An algorithm finding best matches in logarithmic expected time. ACM Trans. Math. Software 3, 209–226Google Scholar
  8. 8.
    Perez, J.C., Vidal, E.: An Approximate Nearest Neighbours Search Algorithm Based on the Extended General Spacefilling Curves Heuristic. In: Amin, A., Pudil, P., Dori, D. (eds.) SPR 1998 and SSPR 1998. LNCS, vol. 1451, pp. 697–706. Springer, Heidelberg (1998)CrossRefGoogle Scholar
  9. 9.
    Perez-Cortes, J.C., Arlandis, J., Llobet, R.: Fast and accurate handwritten character recognition using approximate nearest neighbours search on large databases. In: Amin, A., Pudil, P., Ferri, F., Iñesta, J.M. (eds.) SPR 2000 and SSPR 2000. LNCS, vol. 1876, pp. 767–776. Springer, Heidelberg (2000)CrossRefGoogle Scholar
  10. 10.
    Rosenfeld, A., Pfaltz, J.L.: Sequential Operations in Digital Picture Processing. Journal of the ACM 13(4), 471–494 (1966)MATHCrossRefGoogle Scholar
  11. 11.
    Tomasi, C., Manduchi, R.: Stereo Matching as a Nearest-Neighbor Problem. IEEE Trans. on PAMI 20(3), 333–340 (1998)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2003

Authors and Affiliations

  • Joaquim Arlandis
    • 1
  • Juan-Carlos Perez-Cortes
    • 1
  1. 1.Universitat Politècnica de ValènciaValènciaSpain

Personalised recommendations