A Semi-automatic Methodology for Recognition of Printed Kannada Character Primitives Useful in Character Construction

  • Basavaraj S. Anami
  • Deepa S. GaragEmail author
Conference paper
Part of the Communications in Computer and Information Science book series (CCIS, volume 1037)


Every character of the language having script is written using basic units called primitives. One or more number of primitives connected appropriately result in construction of a character. In the process of character construction identification of primitives is considered as high priority. Through this paper we propose a semi-automatic method for extracting features from primitives for their recognition and further Kannada characters’ construction. The primitives are recognized automatically by adopting the zone features and neighbor classifier. The feature vectors are obtained for all the primitives of Kannada character set and a knowledge base is created. We have used Euclidean distance measure to establish similarity between test input primitives and existing primitives present in the knowledge base for identifying the primitives in Kannada characters. The suggested methodology is tested for 11520 manually extracted primitive images. Average recognition accuracies observed is in the range of 75% to 100% for printed primitives. Application spreads in various verticals of automating literature like calligraphy, digitizing old manuscripts, multimedia teaching, Robot based assistance in handwriting, animation etc.


Kannada printed characters Primitives extraction Classification Zone based features Feature extraction Nearest neighbor K-fold cross validation 


  1. 1.
    Pal, U., Jayadevan, R., Sharma, N.: Handwriting recognition in indian regional scripts: a survey of offline techniques. ACM Trans. Asian Lang. Inf. Process. 11, 1 (2012)CrossRefGoogle Scholar
  2. 2.
    Sharma, O.P., Ghose, M.K., Shah, K.B., Thakur, B.K.: Recent trends and tools for feature extraction in OCR technology. Int. J. Soft Comput. Eng. 2, 220–223 (2013)Google Scholar
  3. 3.
    Santosh, K.C.: Character recognition based on DTW-Radon. In: IAPR, International Conference on Document Analysis and Recognition (ICDAR), pp. 264–268, IEEE, September 2011Google Scholar
  4. 4.
    Santosh, K.C., Wendling, L.: Character recognition based on non-linear multi-projection profiles measure. Front. Comput. Sci. 9(5), 678–690 (2015)CrossRefGoogle Scholar
  5. 5.
    Sheshadri, K., Ambekar, P.K.T., Prasad, D.P., Kumar, R.P.: An OCR system for Printed Kannada using k-means clustering (2010)Google Scholar
  6. 6.
    Nithya, E., Babu, R.: OCR system for complex Printed Kannada characters. Int. J. Adv. Res. Comput. Sci. Softw. Eng. 3, 102–105 (2013)Google Scholar
  7. 7.
    Ragha, L.R., Sasikumar, M.: Using moments features from Gabor directional images for Kannada handwriting character recognition. In: International Conference and Workshop on Emerging Trends in Technology (ICWET 2010) (2010)Google Scholar
  8. 8.
    Sangame, S.K., Ramteke, R.J., Yogesh, V.G.: Recognition of isolated handwritten Kannada characters using invariant moments and chain code. World J. Sci. Technol. 1, 115–120 (2011)Google Scholar
  9. 9.
    Rajput, G.G., Horakeri, R.: Shape descriptors based handwritten character recognition engine with application to Kannada characters. In: International Conference on Computer & Communication Technology (ICCCT) (2011)Google Scholar
  10. 10.
    Rajput, G.G., Horakeri, R.: Zone based handwritten Kannada character recognition using crack code and SVM. In: International Conference on Advances in Computing, Communications and Informatics (ICACCI) (2013)Google Scholar
  11. 11.
    Angadi, S.A., Angadi, S.H.: Structural features for recognition of hand written Kannada character based on SVM. Int. J. Comput. Sci. Eng. Inf. Technol. (IJCSEIT) 5(2), 25–32 (2015)CrossRefGoogle Scholar
  12. 12.
    Dhandra, B.V., Mukarambi, G., Hangarge, M.: Zone based features for handwritten and printed mixed Kannada digits recognition. In: 2011 Proceedings of International Conference on VLSI, Communication & Instrumentation (ICVCI). Int. J. Comput. Appl. (2011)Google Scholar
  13. 13.
    Kumar, K.S.P.: Optical Character Recognition (OCR) for Kannada numerals using left bottom 1/4th segment minimum features extraction. Int. J. Comput. Techol. Appl. 3, 221–225 (2012)Google Scholar
  14. 14.
    Ramappa, M.H., Krishnamurthy, S.: A comparative study of different feature extraction and classification methods for recognition of handwritten Kannada numerals. Int. J. Database Theory Appl. 6, 71–90 (2013)Google Scholar
  15. 15.
    Mamatha, H.R., Srirangaprasad, S., Srikantamurthy, K.: Data fusion based framework for the recognition of Isolated Handwritten Kannada Numerals. Int. J. Adv. Comput. Sci. Appl. (2013)Google Scholar
  16. 16.
    Anami, B.S., Garag, D.S.: Zonal-features based nearest neighbor classification of images of Kannada printed and handwritten vowel and consonant primitives. Glob. J. Comput. Sci. Technol. GJCST 14-F(4) (2014)Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2019

Authors and Affiliations

  1. 1.K.L.E. Institute of TechnologyHubliIndia

Personalised recommendations