Advertisement

Residual Neural Network Vs Local Binary Convolutional Neural Networks for Bilingual Handwritten Digit Recognition

  • Ebrahim Al-wajihEmail author
  • Rozaida Ghazali
  • Yana Mazwin Mohmad Hassim
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 978)

Abstract

Most of the public and government documents in Arabic states are typed or written in bilingual forms; Arabic and Latin languages; such as railway reservation slips, bank withdrawal slips, etc. Using a bilingual system is better than using two systems for every language that is not a practical solution. In this paper, a bilingual digit recognition system is developed using Residual Neural Network (ResNet) and Local Binary Convolutional Neural Networks (LBCNN). The proposed systems are evaluated using a bilingual dataset generated from AHDBase and MNIST datasets. The recognition rate of ResNet or LBCNN is 99.38%. In addition, the proposed systems are applied to MNIST and AHDBase datasets separately. The obtained accuracies for MNIST are 99.27% and 99.51% and for AHDBase are 99.29% and 99.38%, respectively. The resulting performance of ResNet and LBCNN are the highest when they are compared against several state-of-the-art techniques.

Keywords

Bilingual digit recognition Deep learning Residual Neural Network Local binary convolutional neural network 

Notes

Acknowledgment

The authors would like to thank Universiti Tun Hussein Onn Malaysia (UTHM) and Ministry of Education Malaysia for financially supporting this research under the Fundamental Research Grant Scheme (FRGS), Vote No. 1641.

References

  1. 1.
    AlKhateeb JH, Alseid M (2014) DBN-based learning for Arabic handwritten digit recognition using DCT features. In: 2014 6th international conference on computer science and information technology (CSIT). IEEE, pp 222–226Google Scholar
  2. 2.
    Chatterjee A, Malakar S, Sarkar R, Nasipuri M (2018) Handwritten digit recognition using DAISY descriptor: a study. In: 2018 Fifth international conference on emerging applications of information technology (EAIT). IEEE, pp 1–4Google Scholar
  3. 3.
    Lawgali A (2015) Handwritten Digit Recognition based on DWT and DCT. Int J Database Theory Appl 8:215–222CrossRefGoogle Scholar
  4. 4.
    El-Sherif EA, Abdelazeem S (2007) A two-stage system for Arabic handwritten digit recognition tested on a new large database. In: Artificial intelligence and pattern recognition, pp 237–242Google Scholar
  5. 5.
    Mozafari M, Ganjtabesh M, Nowzari-Dalini A, Thorpe SJ, Masquelier T (2018) Bio-inspired digit recognition using spike-timing-dependent plasticity (stdp) and reward-modulated stdp in deep convolutional networks. ArXiv Preprint. https://arxiv.org/abs/180400227
  6. 6.
    Chakraborty S, Paul S, Sarkar R, Nasipuri M (2019) Feature map reduction in CNN for handwritten digit recognition. In: Recent developments in machine learning and data analytics. Springer, Singapore, pp 143–148Google Scholar
  7. 7.
    Siddique F, Sakib S, Siddique MAB (2019) Handwritten digit recognition using convolutional neural network in Python with tensorflow and observe the variation of accuracies for various hidden layersGoogle Scholar
  8. 8.
    Juefei-Xu F, Boddeti VN, Savvides M (2016) Local binary convolutional neural networks. ArXiv Preprint. https://arxiv.org/abs/160806049
  9. 9.
    Das A, Kundu T, Saravanan C (2018) Dimensionality reduction for handwritten digit recognitionGoogle Scholar
  10. 10.
    Kanoun S, Ennaji A, LeCourtier Y, Alimi AM (2002) Script and nature differentiation for Arabic and Latin text images. In: 2002 Proceedings of eighth international workshop on frontiers in handwriting recognition. IEEE, pp 309–313Google Scholar
  11. 11.
    Haboubi S, Maddouri SS, Amiri H (2011) Discrimination between Arabic and Latin from bilingual documents. In: 2011 International conference on communications, computing and control applications (CCCA). IEEE, pp 1–6Google Scholar
  12. 12.
    Al-wajih E, Ahmed M Extended statistical features based on gabor filters for face-based gender classification: classifiers performance comparison. 10Google Scholar
  13. 13.
    Al-wajih E, Ghouti L (2019) Gender recognition using four statistical feature techniques: a comparative study of performance. Evol Intell.  https://doi.org/10.1007/s12065-019-00264-zCrossRefGoogle Scholar
  14. 14.
    Waheeb W, Ghazali R, Deris MM (2015) Content-based sms spam filtering based on the scaled conjugate gradient backpropagation algorithm. In: 2015 12th International conference on fuzzy systems and knowledge discovery (FSKD). IEEE, pp 675–680Google Scholar
  15. 15.
    Lawrence S, Giles CL, Tsoi AC, Back AD (1997) Face recognition: a convolutional neural-network approach. IEEE Trans Neural Netw 8:98–113CrossRefGoogle Scholar
  16. 16.
    LeCun Y (1989) Generalization and network design strategies. Connect Perspect 143–155Google Scholar
  17. 17.
    El-Sawy A, Hazem EB, Loey M (2016) CNN for handwritten arabic digits recognition based on LeNet-5. In: International conference on advanced intelligent systems and informatics. Springer, Cham, pp 566–575Google Scholar
  18. 18.
    LeCun Y (1998) The MNIST database of handwritten digits. Httpyann Lecun ComexdbmnistGoogle Scholar
  19. 19.
    Dhandra BV, Hangarge M, Mukarambi G (2010) Spatial features for handwritten Kannada and English character recognition. IJCA Spec Issue RTIPPR 3:146–151Google Scholar
  20. 20.
    Dhandra BV, Mukarambi G, Hangarge M (2011) A recognition system for handwritten Kannada and English characters. Int J Comput Vis Robot 2:290–301CrossRefGoogle Scholar
  21. 21.
    Dhandra BV, Mukarambi G, Hangarge M (2011) A script independent approach for handwritten bilingual Kannada and Telugu digits recognition. Int J Mach Intell 3Google Scholar
  22. 22.
    Chaudhari SA, Gulati RM (2013) An OCR for separation and identification of mixed English—Gujarati digits using kNN classifier. In: 2013 International conference on intelligent systems and signal processing (ISSP). IEEE, pp 190–193Google Scholar
  23. 23.
    Lehal G, Bhatt N (2000) A recognition system for Devnagri and English handwritten numerals. In: International Conference on Multimodal Interfaces—ICMI 2000, pp 442–449Google Scholar
  24. 24.
    Elgammal AM, Ismail MA (2001) Techniques for language identification for hybrid Arabic-English document images. In: 2001 Proceedings of sixth international conference on document analysis and recognition. IEEE, pp 1100–1104Google Scholar
  25. 25.
    Guo H, Ding X, Zhang Z, Guo F, Wu Y (1995) Realization of a high-performance bilingual Chinese-English OCR system. In: 1995 Proceedings of the third international conference on document analysis and recognition. IEEE, pp 978–981Google Scholar
  26. 26.
    Jawahar CV, Kumar MP, Kiran SR (2003) A bilingual OCR for Hindi-Telugu documents and its applications. In: 2003 Proceedings of seventh international conference on document analysis and recognition. IEEE, pp 408–412Google Scholar
  27. 27.
    Wang K, Jin J, Wang Q (2009) High performance Chinese/English mixed OCR with character level language identification. In: 2009 10th International conference on document analysis and recognition, ICDAR’09. IEEE, pp 406–410Google Scholar
  28. 28.
    Hangarge M, Dhandra BV (2010) Offline handwritten script identification in document images. Int J Comput Appl 4:6–10Google Scholar
  29. 29.
    Pal U, Chaudhuri BB (2001) Automatic identification of English, Chinese, Arabic, Devnagari and Bangla script line. In: 2001 Proceedings of sixth international conference on document analysis and recognition. IEEE, pp 790–794Google Scholar
  30. 30.
    Dhandra BV, Benne RG, Hangarge M (2011) Kannada, Telugu and Devanagari handwritten numeral recognition with probabilistic neural network: a script independent approach. Int J Comput Appl 26:11–16Google Scholar
  31. 31.
    Singhal V, Navin N, Ghosh D (2003) Script-based classification of hand-written text documents in a multilingual environment. In: 2003 Proceedings of 13th international workshop on research issues in data engineering: multi-lingual information management, RIDE-MLIM 2003. IEEE, pp 47–54Google Scholar
  32. 32.
    Rajput GG, Anita HB (2010) Handwritten script recognition using DCT and wavelet features at block level. IJCA Spec Issue RTIPPR 3:158–163Google Scholar
  33. 33.
    He K, Zhang X, Ren S, Sun J (2016) Identity mappings in deep residual networks. In: European conference on computer vision. Springer, Cham, pp 630–645CrossRefGoogle Scholar
  34. 34.
    LeCun Y, Bottou L, Bengio Y, Haffner P (1998) Gradient-based learning applied to document recognition. Proc IEEE 86:2278–2324CrossRefGoogle Scholar
  35. 35.
    Ojala T, Pietikainen M, Harwood D (1994) Performance evaluation of texture measures with classification based on Kullback discrimination of distributions. In: 1994 Proceedings of the 12th IAPR international conference on pattern recognition, conference a: computer vision & image processing, vol 1. IEEE, pp 582–585Google Scholar

Copyright information

© Springer Nature Switzerland AG 2020

Authors and Affiliations

  • Ebrahim Al-wajih
    • 1
    • 2
    Email author
  • Rozaida Ghazali
    • 1
  • Yana Mazwin Mohmad Hassim
    • 1
  1. 1.Faculty of Computer Science and Information TechnologyUniversiti Tun Hussein Onn MalaysiaParit RajaMalaysia
  2. 2.Society Development & Continuing Education CenterHodeidah UniversityAlduraihimiYemen

Personalised recommendations