Advertisement

A Combined Architecture Based on Artificial Neural Network to Recognize Kannada Vowel Modifiers

  • Siddhaling UrolaginEmail author
Conference paper
Part of the Communications in Computer and Information Science book series (CCIS, volume 1037)

Abstract

The document image analysis for Indian scripts such as Kannada poses many challenges due to particular characteristics of the script. The Kannada script has huge set of characters consists of vowels, consonants, consonant conjuncts. The character set also includes compound characters which are formed using the basic symbols. A typical procedure to perform Kannada character recognition is to segment words and characters from the document then carry out recognition. But Kannada has the larger character set, and such an approach will have many classes to recognize. Another method is to segment the character into basic symbols and then perform recognition of the basic symbols. The glyph corresponding to a Kannada character has mainly two parts: consonant and vowel modifiers. This paper, a combined architecture is proposed to perform recognition of Kannada vowel modifiers. Gabor filters are employed to carry out precise segmentation of character into basic symbols. A combined architecture using K-Mean clustering and Artificial Neural Network is developed to recognize segmented vowel modifiers. The 10-fold cross validation is performed and an overall recognition rate of 95.04% is observed.

Keywords

Combined classifier Vowel recognition Kannada document image analysis 

References

  1. 1.
    Hidayatullah, P., Syakrani, N., Suhartini, I., Muhlis, W.: Optical character recognition improvement for license plate recognition in Indonesia. In Sixth UKSim/AMSS European Symposium on Computer Modeling and Simulation (2012)Google Scholar
  2. 2.
    Ramadijanti, N., Basuki, A., Agrippina, G.J.W.: Designing mobile application for retrieving book information using optical character recognition. In: International Conference on Knowledge Creation and Intelligent Computing (KCIC) (2016)Google Scholar
  3. 3.
    Adam, S., Ogier, J.M., Cariou, C., Mullot, R., Labiche, J., Gardes, J.: Symbol and character recognition: application to engineering drawings. Int. J. Doc. Anal. Recognit. 3(2), 89–101 (2000)CrossRefGoogle Scholar
  4. 4.
    Ramiah, S., Liong, T.Y., Jayabalan, M.: Detecting text based image with optical character recognition for English translation and speech using Android. In: IEEE Student Conference on Research and Development (SCOReD) (2015)Google Scholar
  5. 5.
    Pati, P.B., Ramakrishnan, A.G.: OCR in Indian scripts: a survey. J. IETE Tech. Rev. 22(3), 217–227 (2015)CrossRefGoogle Scholar
  6. 6.
    Pal, U., Choudhuri, B.B.: Indian script character recognition: a survey. Pattern Recognit. 37, 1887–1899 (2004)CrossRefGoogle Scholar
  7. 7.
    Ashwin, T.V., Sastry, P.S.: A font and size independent OCR system for printed Kannada documents using support vector machines. J. Sadhana 27(1), 35–58 (2002)CrossRefGoogle Scholar
  8. 8.
    VijayKumar, B., Ramakrishnan, A.G.: Radial basis function and subspace approach for printed Kannada text recognition. In: Proceedings of IEEE International Conference on Acoustics, Speech, and Signal Processing, vol. 5, pp. 321–324 (2004)Google Scholar
  9. 9.
    Nagabhushan, P., Pai, R.M.: Modified region decomposition method and optimal depth decision tree in the recognition of non-uniform sized characters - an experimentation with Kannada characters. Pattern Recognit. Lett. 37, 1467–1475 (1999)CrossRefGoogle Scholar
  10. 10.
    Kunte, R.S., Samuel, R.S.: Wavelet descriptors for recognition of basic symbols in printed Kannada text. Int. J. Wavelets Multiresolution Inf. Process. 5(2), 351–367 (2006)CrossRefGoogle Scholar
  11. 11.
    Hangarge, M., Patil, S., Dhandra, B.V.: Multi-font/size Kannada vowels and numerals recognition based on modified invariant moments. IJCA Spec. Issue Recent. Trends Image Process. Pattern Recognit. Part 2, 126–130 (2010)Google Scholar
  12. 12.
    Kunte, R.S. and Samuel, R.S.: An OCR system for printed Kannada text using two-stage multi-network classification approach employing wavelet features. In: Proceedings of International Conference on Computational Intelligence and Multimedia Applications, pp. 349–353. IEEE Computer Society Press (2007)Google Scholar
  13. 13.
    Aradhya, V.N.M., Kumar, G.H., Noushath, S.: Multilingual OCR system for south Indian scripts and English documents: an approach based on Fourier transform and principal component analysis. Eng. Appl. Artif. Intell. 21(4), 658–668 (2008)CrossRefGoogle Scholar
  14. 14.
    Sagar, B.M., Shobha, G., Ramakanth Kumar, P.: OCR for printed Kannada text to machine editable format using database approach. J. WSEAS Trans. Comput. Arch. 7(6), 766–769 (2008)Google Scholar
  15. 15.
    Sagar, B.M., Shobha, G., Kumar, P.R.: Complete Kannada optical character recognition with syntactical analysis of the script. In: 2008 International Conference on Computing, Communication and Networking (2008)Google Scholar
  16. 16.
    Indira, K., Sethu Selvi, S.: Kannada character recognition system: a review. InterJRI Sci. Technol. 1(2), 30–42 (2009)Google Scholar
  17. 17.
    Tonazzini, A., Bedini, L., Salerno, E.: IJDAR 7, 17 (2004).  https://doi.org/10.1007/s10032-004-0121-8CrossRefGoogle Scholar
  18. 18.
    Urolagin, S., Prema, K.V., Krishna, R.J., Reddy, N.S.: Segmentation of inflected top portions of kannada characters using gabor filters. In: IEEE Second International Conference on Emerging Applications of Information Technology, Kolkata, India, 18–20 Feb 2011, pp. 110–113 (2011).  https://doi.org/10.1109/EAIT.2011.66
  19. 19.
    Urolagin, S., Prema, K.V., Reddy, N.S.: A Gabor filters based method for segmenting inflected characters of kannada script. In: IEEE proceeding of International Conference on Industrial and Information System, pp. 108–113 (2010)Google Scholar
  20. 20.
    Mehrotra, R., Namuduri, K.R., Ranganathan, N.: Gabor filter-based edge detection. Pattern Recognit. 25(12), 1479–1494 (1992)CrossRefGoogle Scholar
  21. 21.
    Chen, J., Sato, Y., Tamura, S.: Orientation space filtering for multiple orientation line segmentation. IEEE Trans. Pattern Anal. Mach. Intell. (PAMI) 22(5), 417–429 (2000)CrossRefGoogle Scholar
  22. 22.
    Kamarainen, J.K., Kyrki, V., Kalviainen, H.: Noise tolerant object recognition using Gabor filtering. In: Proceeding of 14th International Conference on Digital Signal Processing, vol. 2, pp. 1349–1352 (2002)Google Scholar
  23. 23.
    Kamarainen, J.K., Kyrki, V., Kalviainen, H.: Fundamental frequency Gabor filters for object recognition. In: Proceeding of 16th International Conference on Pattern Recognition (ICPR), Quebec, Canada, vol. 1, pp. 628–631 (2002)Google Scholar
  24. 24.
    Bovik, A.C., Clark, M., Geisler, W.S.: Multichannel texture analysis using localized spatial filters. IEEE Trans. Pattern Anal. Mach. Intell. (PAMI) 12(1), 55–73 (1990)CrossRefGoogle Scholar
  25. 25.
    Jain, K., Farrokhnia, F.: Unsupervised texture segmentation using Gabor filters. Pattern Recognit. 24(12), 1167–1186 (1991)CrossRefGoogle Scholar
  26. 26.
    Yoshimura, H., Etoh, M., Kondo, K., Yokoya, N.: Gray-scale character recognition by Gabor jets projection. In: Proceedings of 15th International Conference on Pattern Recognition, vol. 2, pp. 335–338 (2000)Google Scholar
  27. 27.
    Urolagin, S., Prema, K.V., Subba Reddy, N.V.: Robust object recognition using binarized Gabor features under noise and illumination changes. Int. J. Inf. Process. 2(2), 170–182 (2008)Google Scholar
  28. 28.
    Moreno, P., Bernardino, A., Santos-Victor, J.: Gabor parameter selection for local feature detection. In: Proceeding of 2nd Iberian Conference on Pattern Recognition and Image Analysis, pp. 11–19. IBPRIA, Estoril, Portugal (2005)CrossRefGoogle Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2019

Authors and Affiliations

  1. 1.Department of Computer ScienceBirla Institute of Technology and Science, Pilani-DubaiDubaiUAE

Personalised recommendations