Skip to main content
Springer Nature Link
Log in

A Two-Stage Deep Feature Selection Method for Online Handwritten Bangla and Devanagari Basic Character Recognition

  • Original Research
  • Published:
SN Computer Science Aims and scope Submit manuscript

Abstract

In this paper, we have proposed a two-stage deep feature selection (FS) approach for the recognition of online handwritten Bangla and Devanagari basic characters. At the beginning of the approach, we have checked the performance of nine pre-trained transfer learning models namely, DenseNet121, EfficientNetB0, NASNetMobile, VGG-16, VGG-19, ResNet50, InceptionV3, Xception, and MobileNetV2 for the recognition of the said handwritten characters. After that we have considered the best-performing model which is VGG-19 in our case. The obtained features from this model are then reduced using a two-stage FS approach. In the first stage, features are ranked using a filter method, called ReliefF. Then in the second stage, the ranked features are optimized by applying a nature-inspired meta-heuristic, called gray wolf optimization. The experimental outcomes reveal that not only the proposed approach reduces the feature dimension by 88.5% for Bangla and 91.9% for Devanagari, but also it increases classification accuracy for Bangla (reaches 100%) and retains it for Devanagari (which is 99.61%).

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

Access this article

Log in via an institution

Subscribe and save

Springer+ Basic
$34.99 /Month
  • Get 10 units per month
  • Download Article/Chapter or eBook
  • 1 Unit = 1 Article or 1 Chapter
  • Cancel anytime
Subscribe now

Buy Now

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

Similar content being viewed by others

References

  1. Bahlmann C, Burkhardt H. The writer independent online handwriting recognition system frog on hand and cluster generative statistical dynamic time warping. IEEE Trans Pattern Anal Mach Intell. 2004;26(3):299–310.

    Article  Google Scholar 

  2. Farha M, Srinivasa G, Ashwini AJ, Hemant K. Online handwritten character recognition. Int J Comput Sci. 2013;11(5):30–6.

    Google Scholar 

  3. Tappert CC, Suen CY, Wakahara T. The state of online handwriting recognition. IEEE Trans Pattern Anal Mach Intell. 1990;12(8):787–807.

    Article  Google Scholar 

  4. Bawa RK, Rani R. A preprocessing technique for recognition of online handwritten Gurmukhi numerals. In: International Conference on High Performance Architecture and Grid Computing, p. 275–281, 2011.

  5. Gupta M, Gupta N, Agrawal R. Recognition of online handwritten Gurmukhi strokes using support vector machine. In: Proceedings of \(7{\rm th}\) International Conference on Bio-Inspired Computing: Theories and application, p. 495–506, 2012.

  6. Sachan MK, Lehal Singh G, Jain VK. A novel method to segment online Gurmukhi script. In: International Conference on Information Systems for Indian Languages, p. 1–8, 2011.

  7. Sharma A, Kumar R, Sharma RK. HMM-based online handwritten Gurmukhi character recognition. Int J Mach Graph Vis. 2010;19(4):439–49.

    Google Scholar 

  8. Sen S, Bhattacharyya A, Sarkar R, Roy K. BYANJON: a ground truth preparation system for online handwritten Bangla documents. ACM Trans Asian Low-Resour Lang Inf Process. 20(6) (2021).

  9. Swethalakshmi H, Jayaraman A, Chakravarthy VS, Shekhar CC. Online handwritten character recognition of Devanagari and Telugu characters using support vector machines. In: Proceedings of the \(10{\rm th}\) International Workshop on Frontire Handwriting Recognition, p. 367–72, 2006.

  10. Connell SD, Sinha RMK, Jain AK. Recognition of unconstrained online Devanagari characters. In: Proceedings of the \(15{\rm th}\) International Conference on Pattern Recognition, p. 368–71, 2000.

  11. Kubatur S, Sid-Ahmed M, Ahmadi M. A neural network approach to online Devanagari handwritten character recognition. In: Proceedings of the International Conference on High Performance Computing and Simulation, p. 209–14, 2012.

  12. Kumar A, Bhattacharya S. Online Devanagari isolated character recognition for the iPhone using hidden Markov models. In: Proceedings of the International Conference on Students’ Technology Symposium, p. 300–4, 2010.

  13. Santosh KC, Nattee C, Lamiroy B. Relative positioning of stroke-based clustering: a new approach to online handwritten Devanagari character recognition. Int J Image Graph. 2012;12(2):25.

    Article  MathSciNet  Google Scholar 

  14. Parui SK, Guin K, Bhattacharya U, Chaudhuri BB. Online handwritten Bangla character recognition using HMM. In: International Conference on Pattern Recognition, p. 1–4, 2008.

  15. Roy K. Stroke-database design for online handwriting recognition in Bangla. Int J Mod Eng Res. 2012:2534–40.

  16. Bhattacharya U, Gupta BK, Parui SK. Direction code based features for recognition of online handwritten characters of Bangla. In: International Conference on Document Analysis and Recognition, p. 58–62, 2007.

  17. Sen S, Bhattacharyya A, Singh PK, Sarkar R, Roy K, Doermann D. Application of structural and topological features to recognize online handwritten Bangla characters. ACM Trans Asian Low-Resour Lang Inf Process. 2018;17(3):1–19.

    Article  Google Scholar 

  18. Chakraborty R, Mukherjee D, Bhattacharyya A, Mukherjee M, Sur MK, Sen S, Roy K. Online handwritten Bangla and Devanagari character recognition by using CNN: a deep learning concept. In: IEEE 1st International Conference for Convergence in Engineering (ICCE), 2020.

  19. https://towardsdatascience.com/why-deep-learning-is-needed-over-traditional-machine-learning-1b6a99177063. Last accessed 2020/11/03.

  20. Krizhevsky A, Sutskever I, Hinton GE. ImageNet classification with deep convolutional neural networks. Commun ACM. 2012;60(6):84–90.

    Article  Google Scholar 

  21. Sen S, Shaoo D, Paul S, Sarkar R, Roy K. Online handwritten Bangla character recognition using CNN: a deep learning approach. Adv Intell Syst Comput 2018:413–420.

  22. Pham DV. Online handwriting recognition using multi convolution neural networks. Lect Notes Comput Sci LNCS. 2012;7673:310–9.

    Article  Google Scholar 

  23. Baldominos A, Saez Y, Isasi P. Evolutionary convolutional neural networks: an application to handwriting recognition. Int J Neurocomput. 2018;283:38–52.

    Article  Google Scholar 

  24. Deng J, Dong W, Socher R, Li L, Li K, Fei-Fei L. ImageNet: a large-scale hierarchical image database. In: IEEE Conference on Computer Vision and Pattern Recognition, p. 248–55, 2009.

  25. Mehrotra K, Jetley S, Deshmukh A, Belhe S. Unconstrained handwritten Devanagari character recognition using convolutional neural networks. In: Proceedings of the \(4{\rm th}\) International Workshop on Multilingual OCR, p. 1–5, 2013.

  26. Chatterjee S, Dutta R, Ganguly D, Chatterjee K, Roy S. Bengali handwritten character classification using transfer learning on deep convolutional neural network. 2019. arXiv preprint arXiv:1902.11133.

  27. Khan AH, Sarkar SS, Mali K, Sarkar R. A genetic algorithm based feature selection approach for microstructural image classification. Exp Tech. 2021.

  28. Sen S, Saha S, Chatterjee S, Mirjalili S, Sarkar R. A bi-stage feature selection approach for COVID-19 prediction using chest CT images. Appl Intell. 2021.

  29. Ghosh M, Sen S, Maulik U, Sarkar R. Quantum squirrel inspired algorithm for gene selection in methylation and expression data of prostate cancer. In: Applied soft computing. Amsterdam: Elsevier; 2021.

    Google Scholar 

  30. Saha S, Ghosh M, Ghosh S, Sen S, Singh PK, Geem ZW, Sarkar R. Feature selection for facial emotion recognition using cosine similarity-Based harmony search algorithm. Appl Sci. 2020;10(8):2816.

    Article  Google Scholar 

  31. Sen S, Mitra M, Bhattacharyya A. Feature Selection for recognition of online handwritten Bangla characters. Neural Process Lett. 2019;50:2281–304.

    Article  Google Scholar 

  32. De Stefano C, Fontanella F, Marrocco C, di Freca AS. A GA-based feature selection approach with an application to handwritten character recognition. Pattern Recognit Lett. 2014;35:130–41.

    Article  Google Scholar 

  33. Liwicki M, Bunke H. Feature selection for HMM and BLSTM based handwriting recognition of whiteboard notes. Int J Pattern Recognit Artif Intell. 2009;23:907–23.

    Article  Google Scholar 

  34. Kimura Y, Suzuki A, Odaka K. Feature selection for character recognition using genetic algorithm. In: International Conference on Innovative Computing, Information and Control (ICICIC), Kaohsiung, p. 401–4, 2009.

  35. Oliveira LS, Benahmed N, Sabourin R, Bortolozzi F, Suen CY. Feature subset selection using genetic algorithms for handwritten digit recognition. In: Proceedings of 14th Brazilian Symposium on Computer Graphics and Image Processing, Florianopolis, Brazil, p. 362–9, 2001.

  36. Oliveira LS, Sabourin R, Bortolozzi F, Suen CY. Feature selection using multi-objective genetic algorithms for handwritten digit recognition. In: Proceedings of 16th International Conference on Pattern Recognition, vol. 1, p. 568–71, 2002.

  37. Chandrashekar G, Sahin F. A survey on feature selection methods. J Comput Electr Eng. 2014;40:16–28.

    Article  Google Scholar 

  38. Mirjalili S, Mirjalili SM, Lewis A. Grey wolf optimizer. Adv Eng Softw. 2014;69:46–61.

    Article  Google Scholar 

  39. Emary E, Zawbaa HM, Grosan C, Hassenian AE. Feature subset selection approach by gray-wolf optimization. In: Abraham A, Krömer P, Snasel V. (Eds.) Afro-European Conference for Industrial Advancement. Advances in Intelligent Systems and Computing, vol. 334. Springer; 2015.

  40. Yusof Y, Mustaffa Z. Time series forecasting of energy commodity using grey wolf optimizer. In: Proceedings of the International Multi Conference of Engineers and Computer Scientists, Hong Kong, 2015.

  41. Dhargupta S, Ghosh M, Mirjalili S, Sarkar R. Selective opposition based grey wolf optimization. Expert Syst Appl. 2020;151:113–389.

    Article  Google Scholar 

  42. Malakar S, Ghosh M, Chatterjee A, Bhowmik S, Sarkar R. Offline music symbol recognition using daisy feature and quantum grey wolf optimization based feature selection. In: Multimedia tools and applications. Berlin: Springer; 2020.

    Google Scholar 

  43. Emary E, Zawbaa HM, Grosan C. Experienced gray wolf optimization through reinforcement learning and neural networks. IEEE Trans Neural Netw Learn Syst. 2018;29(3):681–94.

    Article  MathSciNet  Google Scholar 

  44. Guha R, Chatterjee B, Sk KH, Ahmed S, Bhattacharya T, Sarkar R. Py_FS: a python package for feature selection using meta-heuristic optimization algorithms. In: Springer AISC series of 3rd International Conference on Computational Intelligence in Pattern Recognition (CIPR-2021), Kolkata, India, 2021.

  45. Sandler M, Howard A, Zhu M, Zhmoginov A, Chen LC. MobileNetV2: inverted residuals and linear bottlenecks. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Salt Lake City, UT, USA, p. 4510–20, 2018.

  46. Russakovsky O, Deng J, Su H, Krause J, Satheesh S, Ma S, Huang Z, Karpathy A, Khosla A, Bernstein M. ImageNet large scale visual recognition challenge. Int J Comput Vis (IJCV). 2015;115:211–52.

    Article  MathSciNet  Google Scholar 

  47. Robnik-Sikonja M, Kononenko I. Theoretical and empirical analysis of Relief and ReliefF. Mach Learn. 2003;53:23–69.

    Article  Google Scholar 

  48. Wang Z, Zhang Y, Chen Z, Yang H, Sun Y, Kang J, Yang Y, Liang X. Application of ReliefF algorithm to selecting feature sets for classification of high resolution remote sensing image. Int Geosci Remote Sens Symp. 2016;2016(41271013):755–8.

    Google Scholar 

  49. Kira K, Rendell LA. A practical approach to feature selection. In: Proceedings of the Ninth International Workshop on Machine Learning, p. 249–56, 1992.

  50. Fehmi Burcin O. Artificial search agents with cognitive intelligence for binary optimization problems. Comput Ind Eng. 2019;136:18–30.

    Article  Google Scholar 

Download references

Funding

One of the authors would like to thank SERB (Grant no. EEQ/2018/000797), DST for financial support in the form of project.

Author information

Authors and Affiliations

  1. Future Institute of Engineering and Management, Kolkata, India

    Ankan Bhattacharyya, Rajatsubhra Chakraborty & Soumyajit Saha

  2. University of Engineering and Management, Kolkata, India

    Shibaprasad Sen

  3. Jadavpur University, Kolkata, India

    Ram Sarkar

  4. West Bengal State University, Kolkata, India

    Kaushik Roy

Authors
  1. Ankan Bhattacharyya
    View author publications

    You can also search for this author inPubMed Google Scholar

  2. Rajatsubhra Chakraborty
    View author publications

    You can also search for this author inPubMed Google Scholar

  3. Soumyajit Saha
    View author publications

    You can also search for this author inPubMed Google Scholar

  4. Shibaprasad Sen
    View author publications

    You can also search for this author inPubMed Google Scholar

  5. Ram Sarkar
    View author publications

    You can also search for this author inPubMed Google Scholar

  6. Kaushik Roy
    View author publications

    You can also search for this author inPubMed Google Scholar

Corresponding author

Correspondence to Kaushik Roy.

Ethics declarations

Conflict of interest

The authors declare that there is no conflict of interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

This article is part of the topical collection “Progresses in Image Processing” guest edited by P. Nagabhushan, Peter Peer, Partha Pratim Roy and Satish Kumar Singh.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Bhattacharyya, A., Chakraborty, R., Saha, S. et al. A Two-Stage Deep Feature Selection Method for Online Handwritten Bangla and Devanagari Basic Character Recognition. SN COMPUT. SCI. 3, 260 (2022). https://doi.org/10.1007/s42979-022-01157-2

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1007/s42979-022-01157-2

Keywords