Indexing Large Class Handwritten Character Database

Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 235)


This paper proposes a method of indexing handwritten characters of a large number of classes by the use of Kd-tree. The Ridgelets and Gabor features are used for the purpose of representation. A multi dimensional feature vectors are further projected to a lower dimensional feature space using PCA. The reduced dimensional feature vectors are used to index the character database by Kd-tree. In a large class OCR system, the aim is to identify a character from a large class of characters. Interest behind this work is to have a quick reference to only those potential characters which can have a best match for given unknown character to be recognized without requiring scanning of the entire database. The proposed method can be used as a supplementary tool to speed up the task of identification. The proposed method is tested on handwritten Kannada character database consisting of 2000 images of 200 classes. Experimental results show that the approach yields a good Correct Index Power (CIP) and also depicts the effectiveness of the indexing approach.


Ridgelet Transform Gabor Transform Kd-tree Handwritten Character Indexing 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Candes, E.J., Donoho, D.L.: Ridgelets: a key to higher-dimensional intermittency? Phil. Trans. R. Soc. Lond. A, 2495–2509 (1999)Google Scholar
  2. 2.
    Do, M.N., Vetterli, M.: Finite ridgelet transform for image representation. IEEE Transactions on Image Processing (2002)Google Scholar
  3. 3.
    Jain, A.K., Prabhakar, S., Hang, L.: A multichannel approach of fingerprint classification. IEEE Transactions on PAMI 21(4), 349–359 (1999)CrossRefGoogle Scholar
  4. 4.
    Jain, A.K., Reisman, J.: A hybrid fingerprint matcher. Pattern Recognition 36(7), 1661–1673 (2003)CrossRefGoogle Scholar
  5. 5.
    Jayaraman, U., Prakash, S., Gupta, P.: Indexing multimodal biometric databases using kd-tree with feature level fusion. In: ICISS 2008, pp. 221–234 (2008)Google Scholar
  6. 6.
    Lu, H., Ooi, B.C., Shen, H.T., Xue, X.: Hierarchical indexing structure for efficient similarity search in video retrieval. IEEE Trans. on Knowledge and Data Engineering 18, 1544–1559 (2006)CrossRefGoogle Scholar
  7. 7.
    Mukherjee, R.: Indexing techniques for fingerprint and iris databases. Master Thesis, West Virginia University (2007)Google Scholar
  8. 8.
    Nagasundara, K.B., Guru, D.S., Manjunath, S.: Indexing of online signatures. International Journal of Machine Intelligence 3, 289–294 (2011)Google Scholar
  9. 9.
    Naveena, C., Manjunath Aradhya, V.N.: An impact of ridgelet transform in handwritten recognition: A study on very large dataset of kannada script. In: IEEE World Congress on Information and Communication Technologies (WCIT), pp. 622–625 (2011)Google Scholar
  10. 10.
    Naveena, C., Manjunath Aradhya, V.N., Niranjan, S.K.: The study of different similarity measure techniques in recognition of handwritten characters. In: ACM International Conference on Advances in Computing, Communications and Informatics (ICACCI), pp. 781–787 (2012)Google Scholar
  11. 11.
    Samet, H.: The Design and Analysis of Spatial Data Structures. Addison-Wesley (1990)Google Scholar
  12. 12.
    Tokas, R., Bhadu, A.: A comparative analysis of feature extraction techniques for handwritten character recognition. International Journal of Advanced Technology Engineering Research (IJATER) 2(4), 215–219 (2012)Google Scholar
  13. 13.
    Turk, M., Pentland, A.: Eigenfaces for recognition. Journal of Cognitive Neuroscience 3, 71–86 (1991)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  1. 1.Department of Studies in Computer ScienceUniversity of MysoreMysoreIndia
  2. 2.Department of Master of Computer ApplicationsSri Jayachamarajendra College of EngineeringMysoreIndia

Personalised recommendations