eCS: Enhanced Character Segmentation – A Structural Approach for Handwritten Kannada Scripts

  • C. Naveena
  • V. N. Manjunath Aradhya
  • S. K. Niranjan
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8284)

Abstract

To build an efficient OCR system, preprocessing task of segmentation process should be in accurate way. In segmentation process, character segmentation plays an important role to obtain clear isolated characters. Character segmentation in Kannada word is a crucial task due to the presence of bottom extension characters (called as Vatthus in Kannada and as extra modifiers in English) and Modifiers. Due to the presence of modifiers and few cursive form of characters the script becomes semi-cursive while writing. With this nature some of the letters are touching each other and also bottom extension characters may get touch to main characters. In this regard, an enhanced Character Segmentation (eCS) approach is proposed for an unconstrained handwritten Kannada scripts. The method is based on thinning, branch point and mixture models. The Expectation-Maximization (EM) algorithm is used to learn the mixture of Gaussians. A cluster mean points are used to estimate the direction and branch point as a reference point for segmenting characters. We experimentally evaluated the proposed method on Kannada words and shown encouraging results.

Keywords

Character Segmentation Thinning Branch Points Mixture Models Kannada Script 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Basu, S., Chaudhuri, C., Kundu, M., Nasipuri, M., Basu, D.K.: Segmentation of online handwritten Bengali script. In: Proceedings of 28th IEEE ACE, pp. 171–174 (2002)Google Scholar
  2. 2.
    Bhowmik, T.K., Roy, A., Roy, U.: Character segmentation for handwritten Bangla words using artificial neural network. In: Proceedings of 1st IAPR TC3NNLDAR (2005)Google Scholar
  3. 3.
    Bishop, C.: Pattern Recognition and Machine Learning. Springer (2006)Google Scholar
  4. 4.
    Garg, N.K., Kaur, L., Jindal, M.K.: The Segmentation of Half Characters in Handwritten Hindi Text. In: Singh, C., Singh Lehal, G., Sengupta, J., Sharma, D.V., Goyal, V. (eds.) ICISIL 2011. Communications in Computer and Information Science, vol. 139, pp. 48–53. Springer, Heidelberg (2011)CrossRefGoogle Scholar
  5. 5.
    Lee, H., Verma, B.: Binary segmentation algorithm for english cursive handwriting recognition. Pattern Recognition 45(4), 1306–1317 (2012)CrossRefGoogle Scholar
  6. 6.
    Maragoudakis, M., Kavallieratou, E., Fakotakis, N.: Improving handwritten character segmentation by incorporating Bayesian knowledge with support vector machines. In: Proceedings of ICASSP 2002, vol. 4, pp. IV-4174 (2002)Google Scholar
  7. 7.
    Mohan, P., Shashikiran, K., Ramakrishnan, A.G.: Unrestricted kannada online handwritten akshara recognition using sdtw. In: ACM - Proceedings of International Workshop on Multilingual OCR (2009)Google Scholar
  8. 8.
    Naveena, C., Manjunath Aradhya, V.N.: Handwritten character segmentation for kannada scripts. In: Proceedings of 2nd World Congress on Information and Communication Technologies (WICT 2012), pp. 144–149 (2012)Google Scholar
  9. 9.
    Pal, U., Belaid, A., Choisy, C.: Touching numeral segmentation using water reservoir concept. Pattern Recognition Letters 24(3), 261–272 (2003)CrossRefGoogle Scholar
  10. 10.
    Prasad, M.M., Sukumar, M., Ramakrishnan, A.G.: Divide and conquer technique in online handwritten kannada character recognition. In: Proceedings of International Conference on Signal Processing and Communications (SPCOM), pp. 1–5 (2010)Google Scholar
  11. 11.
    Sari, T., Souibi, L., Sellami, M.: Off-line handwritten Arabic character segmentation algorithm: ACSA. In: Proceedings of 8th International Workshop on Frontiers in Handwriting Recognition (IWFHR 2002), pp. 452–457 (2002)Google Scholar
  12. 12.
    Sharama, R.K., Singh, A.: Segmentation of handwritten text in Gurumukhi script. International Journal of Image Processing 2(3), 12–17 (2004)Google Scholar
  13. 13.
    Tan, J., Lai, J.H., Wang, C.D., Wang, W.X., Zuo, X.X.: A new handwritten character segmentation method based on nonlinear clustering. Neurocomputing 89(15), 213–219 (2012)CrossRefGoogle Scholar
  14. 14.
    Thungamani, M., Kumar, P.R.: A survey of methods and strategies in handwritten kannada character segmentation. International Journal of Science Research 1(1), 18–23 (2012)Google Scholar
  15. 15.
    Turk, M., Pentland, A.: Eigenfaces for recognition. Journal of Cognitive Neuroscience 3, 71–86 (1991)CrossRefGoogle Scholar
  16. 16.
    Zhao, S., Chi, Z., Shi, P., Yan, H.: Two-stage segmentation of unconstrained handwritten Chinese characters. Pattern Recognition 36(1), 145–156 (2003)CrossRefMATHGoogle Scholar
  17. 17.
    Zhao, X., Chi, Z., Feng, D.: An improved algorithm for segmenting and recognizing connected handwritten characters. In: Proceedings of 11th International Conference on Control, Automation, Robotics and Vision, pp. 1611–1615 (2010)Google Scholar
  18. 18.
    Zheng, Z., Zhao, J., Guo, H., Yang, L., Yu, X., Fang, W.: Character segmentation system based on C# design and implementation. In: Proceedings of 2012 International Workshop on Information and Electronics Engineering (IWIEE). Procedia Engineering, pp. 4073–4078. Elsevier (2012)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2013

Authors and Affiliations

  • C. Naveena
    • 1
  • V. N. Manjunath Aradhya
    • 2
  • S. K. Niranjan
    • 2
  1. 1.Dept. of CSEHKBK College of EngineeringBangaloreIndia
  2. 2.Dept. of Master of Computer ApplicationsSri Jayachamarajendra College of EngineeringMysoreIndia

Personalised recommendations