Online Handwritten Bangla Character Recognition Using CNN: A Deep Learning Approach

  • Shibaprasad Sen
  • Dwaipayan Shaoo
  • Sayantan Paul
  • Ram Sarkar
  • Kaushik Roy
Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 695)


In the present work, a typical Convolutional Neural Network (CNN) architecture has been used for the recognition of online handwritten isolated Bangla characters. A detailed analysis about the effects of using different kernel variations, pooling strategies, and activation functions in the CNN architecture has been performed. In this work, total 10000 character samples have been used and among the samples, 30% have been considered as test set and rest 70% have been used to train the recognition model. On test dataset, the technique has been provided 99.40% recognition accuracy. The outcome is better than some recently proposed handcrafted features used for the recognition of online handwritten Bangla characters.


Online handwriting recognition Feature Bangla script CNN 


  1. 1.
    Kubatur, S., Sid-Ahmed, M., Ahmadi, M.: A neural network approach to online Devanagari handwritten character recognition. In: International Conference on High Performance Computing and Simulation, pp. 209–214 (2012)Google Scholar
  2. 2.
    Swethalakshmi, H., Sekhar, C.C., Chakravarthy, V.S.: Spatiostructural features for recognition of online handwritten characters in Devanagari and Tamil scripts. In: International Conference on Artificial Neural Networks, pp. 230–239 (2007)Google Scholar
  3. 3.
    Swethalakshmi, H., Jayaraman, A., Chakravarthy, V.S., Sekhar, C.C.: On-line handwritten character recognition for Devanagari and Telugu scripts using support vector machines. In: International Workshop on Frontiers in Handwriting Recognition, pp. 367–372 (2006)Google Scholar
  4. 4.
    Tappert, C.C., Suen, C.Y., Wakahara, T.: The state of online handwriting recognition. IEEE Trans. Pattern Anal. Mach. Intell. 787–807 (1990)Google Scholar
  5. 5.
    Vescovo, G.D., Rizzi, A.: Online handwriting recognition by the symbolic histograms approach. In: International Conference on Granular Computing, pp. 686–690 (2007)Google Scholar
  6. 6.
    Bhattacharya, U., Gupta, B.K., Parui, S.K.: Direction code based features for recognition of online handwritten characters of Bangla. In: International Conference on Document Analysis and Recognition, pp. 58–62 (2007)Google Scholar
  7. 7.
    Roy, K.: Stroke-database design for online handwriting recognition in Bangla. Int. J. Mod. Eng. Res. 2534–2540 (2012)Google Scholar
  8. 8.
    Roy, K., Sharma, N., Pal, U.: Online Bangla handwriting recognition system. In: International Conference on Advances in Pattern Recognition, pp. 117–122 (2007)Google Scholar
  9. 9.
    Sen, S.P., Sarkar, R., Roy, K., Hori, N.: Recognize online handwritten Bangla characters using hausdorff distance based feature. In: International Conference on Frontiers in Intelligent Computing: Theory and Application, pp. 541–549 (2017)Google Scholar
  10. 10.
    Sen, S.P., Mitra, M., Chowdhury, S., Sarkar, R., Roy, K.: Quad-tree based image segmentation and feature extraction to recognize online handwritten Bangla characters. In: 7th IAPR TC3 Workshop on Artificial Neural Networks in Pattern Recognition, Ulm, Germany, pp. 246–256 (2016)CrossRefGoogle Scholar
  11. 11.
    Mondal, T., Bhattacharya, U., Parui, S.K., Das, K., Mandalapu, D.: On-line handwriting recognition of Indian scripts—The first benchmark. In: International Conference on Frontiers in Handwriting Recognition, pp. 200–205 (2010)Google Scholar
  12. 12.
    Parui, S.K., Guin, K., Bhattacharya, U., Chaudhuri, B.B.: Online handwritten Bangla character recognition using HMM. In: International Conference on Pattern Recognition, pp. 1–4 (2008)Google Scholar
  13. 13.
    Shin, H.C., Roth, H.R., Gao, M., Lu, L., Xu, Z., Nogues, I., Yao, J., Mollura, D., Summers, R.M.: Deep convolutional neural networks for computer-aided detection: CNN architectures, dataset characteristics and transfer learning. IEEE Trans. Med. Imaging 35, 1285–1298 (2016)CrossRefGoogle Scholar
  14. 14.
    Bai, J., Chen, Z., Feng, B., Xu, B.: Image character recognition using deep convolutional neural network learned from different languages. In: IEEE International Conference on Image Processing, pp. 2560–2564 (2014)Google Scholar
  15. 15.
    Yuan, A., Bai, G., Jiao, L., Liu, Y.: Offline handwritten English character recognition based on convolutional neural network. In: International Workshop on Document Analysis Systems, pp. 125–129 (2012)Google Scholar
  16. 16.
    Bouchain, D.: Character recognition using convolutional neural networks. Inst. Neural Inf. Process. (2007)Google Scholar
  17. 17.
    Acharya, S., Pant, A.K., Gyawali, P.K.: Deep learning based large scale handwritten Devanagari character recognition. In: International Conference on Software, Knowledge, Information Management and Applications (2015)Google Scholar
  18. 18.
    Singh, P., Verma, A., Chaudhari, N.S.: Deep convolutional neural network classifier for handwritten Devanagari character recognition. Inf. Syst. Des. Intell. Appl. 551–561 (2016)Google Scholar
  19. 19.
    Mehrotra, K., Jetley, S., Deshmukh, A., Belhe, S.: Unconstrained handwritten Devanagari character recognition using convolutional neural networks. In: Proceedings of 4th International Workshop on Multilingual OCR (2013)Google Scholar
  20. 20.
    Rahman, M.M., Akhand, M.A.H., Islam, S., Shill, P.C., Rahman, M.H.: Bangla handwritten character recognition using convolutional neural network. Int. J. Image Graph. Signal Process. 8(3), 42–49 (2015)CrossRefGoogle Scholar
  21. 21.
    Maitra, D.S., Bhattacharya, U., Parui, S.K.: CNN based common approach to handwritten character recognition of multiple scripts. In: International Conference on Document Analysis and Recognition, pp. 1021–1025 (2015)Google Scholar
  22. 22.
    Janocha, K., Czarnecki, W.M.: On Loss Functions for Deep Neural Networks in Classification, pp. 1–10 (2017)Google Scholar
  23. 23.
    Oland, A., Bansal, A., Dannenberg, R.B., Raj, B.: Be Careful What You Backpropagate: A Case For Linear Output Activations & Gradient Boosting (2017).
  24. 24.
    Jeong, J.M., Choi, S.J., Kim, C.: Correcting method of numerical errors of soft-max and cross entropy according to CNN’s output value. Neural Comput. Appl. 1(1), 15–20 (2017)Google Scholar

Copyright information

© Springer Nature Singapore Pte Ltd. 2018

Authors and Affiliations

  • Shibaprasad Sen
    • 1
  • Dwaipayan Shaoo
    • 1
  • Sayantan Paul
    • 2
  • Ram Sarkar
    • 2
  • Kaushik Roy
    • 3
  1. 1.Future Institute of Engineering & ManagementKolkataIndia
  2. 2.Jadavpur UniversityKolkataIndia
  3. 3.West Bengal State UniversityKolkataIndia

Personalised recommendations