Skip to main content
SpringerLink
Log in

Bayesian Versus Convolutional Networks for Arabic Handwriting Recognition

  • Research Article - Computer Engineering and Computer Science
  • Published:
Arabian Journal for Science and Engineering Aims and scope Submit manuscript

Abstract

In this paper, the ability of Bayesian and convolutional neural networks (CNNs), as two different machine learning methods, to recognize Arabic handwritten words is analyzed. Our contribution is threefold. First, we describe the main highlights of the dynamic Bayesian network (DBN) architecture, especially when compared to standard Bayesian networks. For that, some structural features are extracted from word image and considered as input for different architectures of Bayesian networks (BNs) such as Naïve Bayes (NB), Tree Augmented Naïve Bayes (TAN), Forest Augmented Naïve Bayes (FAN) and Hidden Markov model (HMM). Features are extracted based on the word baseline which has been estimated to mainly cope with the problems of inclination and distortions. Decisions about word classification are then inferred using multiples models of BNs. Second, we model a deep learning architecture: a CNN that convolves learned features with input data and uses 2D convolutional layers that makes it well suited to 2D word image processing. Third, we compare the behavior of DBN–CNN and propose to combine them to exploit their advantages. Experiments are carried on the standard IFN–ENIT database. The obtained results show the relatively high accuracy of the DBN and CNN combination: 95.20% compared to the remaining models.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11
Fig. 12
Fig. 13
Fig. 14
Fig. 15
Fig. 16
Fig. 17
Fig. 18
Fig. 19
Fig. 20
Fig. 21
Fig. 22
Fig. 23

Similar content being viewed by others

References

  1. Lawrence, R.; Rabiner, R.: A tutorial on hidden Markov models and selected applications in speech recognition. Proc. IEEE 77(2), 257–286 (1989)

    Article  Google Scholar 

  2. Suen, C.Y.; Berthod, M.; Mori, S.S.: Automatic recognition of handprinted characters: the state of the art. Proc. IEEE 68(4), 469–487 (1980)

    Article  Google Scholar 

  3. Biadsy, F.; Saabni, R.; El-Sana, J.: Segmentation-free online Arabic handwriting recognition. IJPRAI 25, 1009–1033 (2011)

    Google Scholar 

  4. El Yacoubi, M.; Gilloux, R.; Sabourin, C.; Suen, Y.: An HMM-based approach for off-line unconstrained handwritten word modelling and recognition. IEEE Trans. PAMI 21(8), 752–760 (1999)

    Article  Google Scholar 

  5. Mangrich, M.V.: Hierarchical Feature Extraction: A Stepwise Approach to Image Classification. The University of Montana, Missoula (2000)

    Google Scholar 

  6. Website of IFN-ENIT database (September 2006). http://www.ifnenit.com

  7. Jayech, K.; Trimech, N.; Mahjoub, M.A.; Essoukri Benamara, N.: Dynamic hierarchical Bayesian network for Arabic handwritten word recognition . In: Proceedings of the 4th International Conference on ICT and Accessibility, Hammamet, pp. 1–6 (2013)

  8. Touj, S.M.; Essoukri Benamara, N.; Amiri, H.: Arabic handwritten words recognition based on a planar hidden Markov model. IAJIT 2(4), 318–325 (2005)

    Google Scholar 

  9. Kessentini, Y.; Paquest, T.; Benhamadou, A.: Off-line handwritten word recognition using multi-stream hidden Markov models. In: Proceedings of the 19th International Conference on Pattern Recognition, Tampa, pp. 1–4 (2008)

  10. AlKhateeb, J.H.; Pauplin, O.; Ren, J.; Jiang, J.: Performance of hidden Markov model and dynamic Bayesian network classifiers on handwritten Arabic word recognition. Knowl. Based Syst. 24(5), 680–688 (2011)

    Article  Google Scholar 

  11. AlKhateeb, J.H.: Offline handwritten Arabic digit recognition using dynamic Bayesian network. In: Proceedings of ICCIT, pp. 176–180 (2012)

  12. Mahjoub, M.A.; Ghanmy, N.; Jayech, K.; Miled, I.: Multiple models of Bayesian networks applied to offline recognition of Arabic handwritten city names. J. Imag. Robot. 9(1) (2013)

  13. Li, H.; Lin, Z.; Shen, X.; Brandt, J.; Hua, G.: A convolutional neural network cascade for face detection. In: Proceedings of IEEE Conference on Computer Vision and Pattern Recognition (CVPR) (2015)

  14. Crstea, B.; Likforman-Sulem, L.: Tied spatial transformer networks for digit recognition. In: ICFHR (2016)

  15. Chengz, C.; Zhangy, X.; Shaoz, X.; Zhouz, X.: Handwritten Chinese character recognition by joint classification and similarity ranking. In: ICFHR (2016)

  16. Jarrett, K.; Kavukcuoglu, K.; Ranzato, M.A.; LeCun, Y.: What is the best multi-stage architecture for object recognition? In: 2009 IEEE 12th International Conference on Computer Vision, pp. 2146–2153 (2009)

  17. Ciresan, D.; Meier, U.; Schmidhuber, J.: Multi-column deep neural networks for image classification. In: 2012 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp. 3642–3649 (2012)

  18. Ranzato, M.A.; Huang, F.J.; Boureau, Y.-L.; LeCun, Y.: Unsupervised learning of invariant feature hierarchies with applications to object recognition. In: Computer Vision and Pattern Recognition, CVPR, pp. 1–8 (2007)

  19. Poultney, C.; Chopra, S.; Cun, Y.L.; et al.: Efficient learning of sparse representations with an energy-based model. In: Advances in Neural Information Processing Systems, pp. 1137–1144 (2006)

  20. Goodfellow, I.J.; Bulatov, Y.; Ibarz, J.; Arnoud, S.; Shet, V.: Multi-digit number recognition from street view imagery using deep convolutional neural networks (2013). arXiv preprint arXiv:1312.6082

  21. Shah, P.; Karamchandani, S.; Nadkar, T.; Gulechha, N.; Koli, K.; Lad, K.: OCR-based chassis-number recognition using artificial neural networks. In: Vehicular Electronics and Safety (ICVES), pp. 31–34 (2009)

  22. Touj, S.; Essoukri Benamara, N.; Amiri, H.: A hybrid approach for off-line Arabic handwriting recognition based on a Planar Hidden Markov modelling. Proceedings of the International Conference on Document Analysis and Recognition, Curitiba, Brazil 2, 964–968 (2007)

    Google Scholar 

  23. AL-Shatnawi, A.; Omar, K.: Methods of Arabic language baseline detection—the State of Art. IJCSNS Int. J. Comput. Sci. Netw. Secur. 8(10) (2008)

  24. Pechwitz, M.; Maergner, V.: Baseline estimation for Arabic handwritten words. In: Frontiers in Handwriting Recognition, pp. 479–484 (2002)

  25. Parhami, B.; Taraghi, M.: Automatic recognition of printed Farsi texts. Pattern Recognit. 14(1–6), 395–403 (1981)

    Article  Google Scholar 

  26. Nawaz, S.N.; Sarfraz, M.; Zidouri, A.; Al-Khatib, W.G.: An approach to offline Arabic character recognition using neural networks. In: 10th IEEE International Conference on Electronics, Circuits and Systems (ICECS), vol. 3, pp. 1328–1331. 14–17 Dec. 2003

  27. Sarfraz, M.; Nawaz, S.N.; Al-Khuraidly, A.: Offline Arabic text recognition system. International Conference on Geometric Modeling and Graphics (GMAG), London, pp. 30–36, 16–18 July 2003

  28. El-Hajj, R.; likforman-Sulem, L.; Mokbe, C.: Arabic handwriting recognition using baseline dependant features and hidden Markov modeling. In: Proceedings of the 2005 Eight International Conference on Document Analysis and Recognition (ICDAR), vol. 20(5), pp. 1520–5263 (2005)

  29. Al-Rashaideh, H.: Preprocessing phase for Arabic Word handwritten recognition. Russ. Acad. Sci. 6(1), 11–19 (2006)

    Google Scholar 

  30. AlKhateeb, J.H.; Ipson, J.S.; Jiang, J.: Knowledge-based baseline detection and optimal thresholding for words segmentation in efficient preprocessing of handwritten Arabic text. In: Fifth International Conference on Information Technology: New Generations, IEEE computer society, pp. 1158–1159 (2008)

  31. Farooq, F.; Govindaraju, V.; Perrone, M.: Preprocessing methods for handwritten Arabic documents (ICDAR). In: Proceedings of the Eight International Conference on Document Analysis and Recognition, pp. 267–271 (2005)

  32. Burrow, P.: Arabic handwriting recognition. M.Sc. thesis, University of Edinburgh, England (2004)

  33. Khémiri, A.; Kacem, A.; Belaïd, A.; Elloumi, M.: Arabic handwritten words off-line recognition based on HMMs and DBNs. In: Proceedings of ICDAR (2015)

  34. McCallum, A.: Hidden Markov models Baum Welch algorithm. In: Introduction to Natural Language Processing CS 585 (2004)

  35. Murphy, K.P.: Dynamic Bayesian networks, representation, inference and learning. Ph.D. dissertation, UC Berkeley, Computer Science Division, July 2002

  36. Hallouli, K.; Likforman-Sulem, L.; Sigelle, M.: A Comparative Study Between Decision Fusion and Data Fusion in Markovian Printed Character Recognition. ICPR, Quebec City (2002)

    Book  Google Scholar 

  37. Kacem, A.; Khémiri, A.; Aouiti, N.; Aouadi, N.: Systeme base de MMC, pour la reconnaissance de noms propres manuscrits Arabes. Proceedings of CIDE, Tunisia, 16–18 Oct. 2012

  38. Khémiri, A.; Kacem, A.; Belaïd, A.: Towards Arabic handwritten word recognition via probabilistic graphical models. Proceedings of ICFHR, 1–4 September, Island Crete (2014)

  39. Kacem, A.; Khémiri,; Belaïd, A.: A PGM-based system for Arabic handwritten word recognition. ELCVIA 13(3), 41–62 (2014)

    Article  Google Scholar 

  40. Wu, J.: Introduction to Convolutional Neural Networks, National Key Lab for Novel Software Technology Nanjing University, China, 1 May 2017

  41. Bae, H.; Monti, S.; Montano, M.; Steinberg, M.H.; Perls, T.T.; Sebastiani, P.: Learning Bayesian networks from correlated data. Sci. Rep. 6, 25156 (2016)

    Article  Google Scholar 

  42. Margner, V.; El Abed, H.: ICDAR 2009 Arabic handwriting recognition competition. In: ICDAR, Barcelona, Spain, pp. 1383–1387, July 2009

  43. Margner, V.; Pechwitz, M.; Abed, H.E.: ICDAR 2005 Arabic handwriting recognition competition. In: ICDAR, Seoul, Korea, vol. 1, pp. 70–74, Aug. 2005

  44. Mrgner, V.; El Abed, H.: ICFHR 2010 Arabic handwriting recognition competition. In: ICFHR, Nov 2010

  45. Natarajan, P.; Saleem, S.; Prasad, R.; MacRostie, E.; Subramanian, K.: Arabic and Chinese handwriting recognition. In: Chapter Multi-lingual Offline Handwriting Recognition Using Hidden Markov Models: A Script-Independent Approach, vol. 4768/2008 of LNCS, pp. 231–250, Springer, Berlin (2008)

  46. Schambach, M.-P.; Rottland, J.; Alary, T.: How to convert a Latin handwriting recognition system to Arabic. In: ICFHR (2008)

Download references

Funding

Funding was provided by Ministèe de l’Enseignement Supérieur, de la Recherche Scientifique et des Technologies de l’Information et de la Communication (TN)

Author information

Authors and Affiliations

  1. ENSIT-LaTICE, University of Tunis, 5 Avenue Taha Husseïn, Babmnara, BP 56, 1008, Tunis, Tunisia

    Akram Khémiri, Afef Kacem Echi & Mourad Elloumi

Authors
  1. Akram Khémiri
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Afef Kacem Echi
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Mourad Elloumi
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Akram Khémiri.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Khémiri, A., Echi, A.K. & Elloumi, M. Bayesian Versus Convolutional Networks for Arabic Handwriting Recognition. Arab J Sci Eng 44, 9301–9319 (2019). https://doi.org/10.1007/s13369-019-03939-y

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s13369-019-03939-y

Keywords

Search

Navigation