Fast and Accurate Handwritten Character Recognition Using Approximate Nearest Neighbours Search on Large Databases

  • Juan C. Pérez-Cortes
  • Rafael Llobet
  • Joaquim Arlandis
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1876)


In this work, fast approximate nearest neighbours search algorithms are shown to provide high accuracies, similar to those of exact nearest neighbour search, at a fraction of the computational cost in an OCR task. Recent studies [26,15] have shown the power of k-nearest neighbour classifiers (k-nn) using large databases, for character recognition. In those works, the error rate is found to decrease consistently as the size of the database increases. Unfortunately, a large database implies large search times if an exhaustive search algorithm is used. This is often cited as a major problem that limits the practical value of the k- nearest neighbours classification method. The error rates and search times presented in this paper prove, however, that k-nn can be a practical technique for a character recognition task.


Handwriting Recognition OCR Fast Nearest Neighbour Search Approximate Search k-NN 


  1. 1.
    S. Arya, D.M. Mount, N.S. Netanyahu, R. Silverman, and A. Wu. An optimal algorithm for approximate nearest neighbor searching. Journal of the ACM, 45:891–923, 1998.zbMATHCrossRefMathSciNetGoogle Scholar
  2. 2.
    N. Beckmann, H.-P. Kriegel, R. Schneider, and B. Seeger. The r*-tree: An efficient and robust access method for points and rectangles. In ACM SIGMOD Conf. on the Management of Data 90, Atlantic City., May 1990.Google Scholar
  3. 3.
    J.S. Beis and D.G. Lowe. Indexing without invariants in 3d object recognition. IEEE Trans. on PAMI, 21(10):1000–1015, October 1999.Google Scholar
  4. 4.
    J.L. Bentley, B.W. Weide, and A.C. Yao. Optimal expected time algorithms for closest point problems. ACM Trans. on Math. Software, 6:563–580, 1980.zbMATHCrossRefMathSciNetGoogle Scholar
  5. 5.
    S. Berchtold, D.A. Keim, and H.P. Kriegel. The x-tree: An index structure for high-dimensional data. In Proc. 22nd Very Large Database Conference, Bombay, India, pages 28–39, 1996.Google Scholar
  6. 6.
    M. Bern. Approximate closest-point queries in high dimensions. Pattern Recognition, 45:95–99, 1993.zbMATHMathSciNetGoogle Scholar
  7. 7.
    S. Brin. Near neighbor search in large metric spaces. In Proc. 21st Inter. Conf. on Very Large Data Bases, pages 574–584, 1995.Google Scholar
  8. 8.
    P. A. Devijver and J. Kittler. On the edited nearest neighbour rule. pages 72–80. Proceedings of the 5th International Conference on Pattern Recognition, IEEE Computer Society Press, Los Alamitos, CA, 1980.Google Scholar
  9. 9.
    L. Devroye, L. Györfi, and G. Lugosi. A Probabilistic Theory of Pattern Recognition. Springer-Verlag., 1996.Google Scholar
  10. 10.
    J. H. Friedman, J. L. Bentley, and R. A. Finkel. An algorithm finding best matches in logarithmic expected time. ACM Trans. Math. Software, 3:209–226, 1977.zbMATHCrossRefGoogle Scholar
  11. 11.
    K. Fukunaga and P. M. Narendra. A branch and bound algorithm for computing k-nearest neighbors. 24:750–753, 1975.zbMATHMathSciNetGoogle Scholar
  12. 12.
    S. Geva and J. Sitte. Adaptive nearest neighbor pattern classification. IEEE Trans on Neural Networks, 2(2):318–322, 1991.CrossRefGoogle Scholar
  13. 13.
    P.J. Grother and G.T. Candela. Comparison of handprinted digit classifiers. In NISTIR, 1993.Google Scholar
  14. 14.
    A. Guttman. R-trees: A dynamic index structure for spatial searching. In UCB, Elec.Res.Lab, Res.R. No.M83-64, with Stonebraker, M., 1983.Google Scholar
  15. 15.
    T.M. Ha and H. Bunke. Off-line, handwritten numeral recognition by perturbation method. IEEE Trans. on PAMI, 19(5):535–539, May 1997.Google Scholar
  16. 16.
    P.E. Hart. The condensed nearest neighbor rule. IEEE Trans. on Information Theory, 125:515–516, 1968.CrossRefGoogle Scholar
  17. 17.
    R. Indyk and R Motwani. Approximate nearest neighbors: Towards removing the curse of dimensionality. In 30th Symposium on Theory of Computing, 1998.Google Scholar
  18. 18.
    B. S. Kim and S. B. Park. A fast k nearest neighbor finding algorithm based on the ordered partition. IEEE Trans. on PAMI, 8:761–766, 1986.zbMATHGoogle Scholar
  19. 19.
    T. Kohonen. Self Organization and Associative Memory. Springer-Verlag., 1988.Google Scholar
  20. 20.
    E. Kushilevitz, R. Ostrovsky, and Y. Rabani. Efficient search for approximate nearest neighbor in high dimensional spaces. In 30th Symposium on Theory of Computing, 1998.Google Scholar
  21. 21.
    G. Loizou and S. J. Maybank. The nearest neighbor and the Bayes error rates. IEEE Trans. on PAMI, 9:254–262, 1987.zbMATHGoogle Scholar
  22. 22.
    L. Miclet and M. Dabouz. Approximative fast nearest neighbor recognition. Pat tern Recognition Letters, 1:277–285, 1983.CrossRefGoogle Scholar
  23. 23.
    J.C. Perez and E. Vidal. An approximate nearest neighbours search algorithm based on the extended general spacefilling curves heuristic. In Workshop on Statistical Pattern Recognition SPR-98, Sydney, Australia., 1998.Google Scholar
  24. 24.
    J.C. Perez and E. Vidal. The extended general spacefilling curves heuristic. In Intl. Conf. on Pattern Recognition ICPR-98, Brisbane, Australia., 1998.Google Scholar
  25. 25.
    H. Robinson. Database Analysis and Design., 1981.Google Scholar
  26. 26.
    S.J. Smith, Sims K. Bourgoin, M.O., and H.L. Voorhees. Handwritten character classification using nearest neighbor in large databases. IEEE Trans. on PAMI, 16(9):915–919, September 1994.Google Scholar
  27. 27.
    R.F. Sproull. Refinements to nearest-neighbor searching in k-dimensional trees. Algorithmica, 6:579–589, 1991.zbMATHCrossRefMathSciNetGoogle Scholar
  28. 28.
    C. Tomasi and R. Manduchi. Stereo matching as a nearest-neighbor problem. IEEE Trans. on PAMI, 20(3):333–340, March 1998.Google Scholar
  29. 29.
    D.L. Wilson. Asymptotic properties of nearest neighbor rules using edited data. IEEE Trans. on Systems, Man and Cybernetics, 2:408–420, 1972.zbMATHCrossRefGoogle Scholar
  30. 30.
    P. Yianilos. Data structures and algorithms for nearest neighbor search in general metric spaces. In 4th ACM Symp. on Discrete Algorithms, pages 311–321, 1993.Google Scholar
  31. 31.
    P. Zakarauskas and J.M. Ozard. Complexity analysis for partitioning nearestneighbor searching algorithms. IEEE Trans. on PAMI, 18(6):663–668, June 1996.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2000

Authors and Affiliations

  • Juan C. Pérez-Cortes
    • 1
  • Rafael Llobet
    • 1
  • Joaquim Arlandis
    • 1
  1. 1.Instituto Tecnológico de InformáticaUniversidad Politécnica de ValenciaValenciaSpain

Personalised recommendations