Multi-script Identification from Printed Words

  • Saumya  JetleyEmail author
  • Kapil Mehrotra
  • Atish Vaze
  • Swapnil Belhe
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 8814)


In today’s multi-script scenario, documents contain page, paragraph, line and up to word level intermixing of different scripts. We need a script recognition approach that can perform well even at the lowest semantically-valid level of words so as to serve as a generic solution. The present paper proposes a combination of Histogram of Oriented Gradients (HoG) and Local Binary Patterns (LBP), extracted over words, to capture the unique and discriminative structural formations of different scripts. Tested over MILE printed-word data set, this concatenated feature descriptor yields a state-of-the-art average recognition accuracy of 97.4 % over a set of 11 Indian scripts.

In an end-to-end document recognition system it is correct to assume a skew correction unit prior to script identification. Depending on the amount of skew, the skew correction unit can either yield a correctly aligned document or an inverted one. For script identification in such scenarios, we introduce novel modifications over existing HoG and LBP features to propose - Inversion Invariant HoG (II-HoG) and Inversion Invariant LBP (II-LBP) in order to achieve text inversion invariance. Once the script is recognized, script-specific HoG and LBP feature combination can be used to find the text alignment i.e. 0° or 180° for correction. For the MILE database, first-level inversion-invariant script-identification accuracy for 11 script-set is 95.8 % (1 % gain over the existing best) while the second-level script-specific orientation-detection accuracy is averaged at 97.7 %.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Pati, P.B., Ramakrishnan, A.G.: Word level multi-script identification. Pattern Recogn. Lett., 1218–1229 (2008)Google Scholar
  2. 2.
    Khoddami, M., Behrad, A.: Farsi and latin script identification using curvature scale space features. In: 2010 10th Symposium on Neural Network Applications in Electrical Engineering (NEUREL), pp. 213–217 (2010)Google Scholar
  3. 3.
    Chanda, S., Pal, U., Franke, K., Kimura, F.: Script identification: A han and roman script perspective. In: 2010 20th International Conference on Pattern Recognition (ICPR), pp. 2708–2711 (2010)Google Scholar
  4. 4.
    Roy, K., Alaei, A., Pal, U.: Word-wise handwritten persian and roman script identification. In: 2010 International Conference on Frontiers in Handwriting Recognition (ICFHR), pp. 628–633 (2010)Google Scholar
  5. 5.
    Chanda, S., Franke, K., Pal, U.: Identification of indic scripts on torn-documents. In: 2011 International Conference on Document Analysis and Recognition (ICDAR), pp. 713–717 (2011)Google Scholar
  6. 6.
    Wang, N., Lam, L., Suen, C.: Noise tolerant script identification of printed oriental and english documents using a downgraded pixel density feature. In: 2010 20th International Conference on Pattern Recognition (ICPR), pp. 2037–2040 (2010)Google Scholar
  7. 7.
    Pan, J., Tang, Y.: A rotation-robust script identification based on bemd and lbp. In: 2011 International Conference on Wavelet Analysis and Pattern Recognition (ICWAPR), pp. 165–170 (2011)Google Scholar
  8. 8.
    Ghosh, S., Chaudhuri, B.: Composite script identification and orientation detection for indian text images. In: 2011 International Conference on Document Analysis and Recognition (ICDAR), pp. 294–298 (2011)Google Scholar
  9. 9.
    Aithal, P., Rajesh, G., Acharya, D., Subbareddy, N.: Text line script identification for a tri-lingual document. In: 2010 International Conference on Computing Communication and Networking Technologies (ICCCNT), pp. 1–3 (2010)Google Scholar
  10. 10.
    Gopakumar, R., Subbareddy, N., Makkithaya, K., Acharya, D.: Zone-based structural feature extraction for script identification from indian documents. In: 2010 International Conference on Industrial and Information Systems (ICIIS), pp. 420–425 (2010)Google Scholar
  11. 11.
    Zhou, L., Ping, X., Zheng, E., Guo, L.: Script identification based on wavelet energy histogram moment features. In: 2010 IEEE 10th International Conference on Signal Processing (ICSP), pp. 980–983 (2010)Google Scholar
  12. 12.
    Das, M., Rani, D., Reddy, C.R.K.: Heuristic based script identification from multilingual text documents. In: 2012 1st International Conference on Recent Advances in Information Technology (RAIT), pp. 487–492 (2012)Google Scholar
  13. 13.
    Hiremath, P., Shivashankar, S., Pujari, J., Mouneswara, V.: Script identification in a handwritten document image using texture features. In: 2010 IEEE 2nd International Advance Computing Conference (IACC), pp. 110–114 (2010)Google Scholar
  14. 14.
    Ma, H., Doermann, D.: Word level script identification for scanned document images. In: Proc. of Int. Conf. on Document Recognition and Retrieval (SPIE), pp. 178–191 (2004)Google Scholar
  15. 15.
    Dalal, N., Triggs, B.: Histograms of oriented gradients for human detection. In: CVPR, pp. 886–893 (2005)Google Scholar
  16. 16.
    Ojala, T., Pietikäinen, M., Harwood, D.: A comparative study of texture measures with classification based on featured distributions. Pattern Recognition 29(1), 51–59 (1996)CrossRefGoogle Scholar
  17. 17.
    Horn, R.A., Johnson, C.R.: Norms for Vectors and Matrices. In: Matrix Analysis. Cambridge University Press (1990)Google Scholar
  18. 18.
    Chang, C.-C., Lin, C.-J.: LIBSVM: A library for support vector machines. ACM Transactions on Intelligent Systems and Technology 2, 27:1–27:27 (2011)CrossRefGoogle Scholar
  19. 19.
    Bloomberg, D.S., Kopec, G.E., Dasari, L.: Measuring document image skew and orientation. In: IS&T/SPIE’s Symposium on Electronic Imaging: Science & Technology. International Society for Optics and Photonics, pp. 302–316 (1995)Google Scholar

Copyright information

© Springer International Publishing Switzerland 2014

Authors and Affiliations

  • Saumya  Jetley
    • 1
    Email author
  • Kapil Mehrotra
    • 1
  • Atish Vaze
    • 1
  • Swapnil Belhe
    • 1
  1. 1.Centre for Development of Advanced Computing (C-DAC)PuneIndia

Personalised recommendations