Abstract
Online handwritten word recognition (OHR) in low-resource languages such as Bangla is still an open problem. Although the need and importance of OHR are increasing nowadays, research works on word-level recognition are few (specifically for Bangla script), and there is a lot of room for improving recognition performance. In the current work, we employed different Recurrent Neural Network (RNN) architectures such as Long Short-Term Memory (LSTM), Bidirectional Long Short-Term Memory (BLSTM), Gated Recurrent Unit (GRU), and Bidirectional Gated Recurrent Unit (BGRU) for the recognition of online handwritten Bangla words written in an unconstrained domain. One of the challenges includes the variable number of strokes used to write words. This study aims to develop a segmentation-free recognition module where the features from constituent strokes of the word sample are fed to the developed RNN architectures. Sequential and dynamic information obtained from the strokes is considered as the features for the current experiment. The customized architecture of BLSTM known as BWordDeepNet (Bangla Word Deep-learning Network) provides the best performance with 98.35% correct recognition accuracy on the dataset having 7992 online handwritten Bangla word samples. Additionally, the model achieves a numerical gain of 8.08% compared to the Bangla word recognition work mentioned in [38] that was performed on the same word dataset containing 5550 word samples. We have also compared the performance of our proposed model with state-of-the-art techniques used for the same purpose.
Similar content being viewed by others
Data Availability
The data used in this experiment will be made available to researcher on request for research purpose
References
Almuallim H, Yamaguchi H (1987) A method of recognition of arabic cursive handwriting. IEEE Trans Pattern Anal & Mach Intell, 5:715–722
Baghshah MS, Shouraki SB, Kasaei S (2006) A novel fuzzy classifier using fuzzy lvq to recognize online persian handwriting. In: 2nd International conference on information and communication technologies, 1:1878—1883
Bai ZL, Huo Q (2005) A study on the use of 8-directional features for online handwritten chinese character recognition. In: Eighth international conference on document analysis and recognition, pp 262—266
Beigi HS, Nathan K, Clary GJ et al (1994) Challenges of handwriting recognition in farsi. In: Arabic and other languages with similar writing styles an on-line digit recognizer. Proceedings of the \(2^{nd}\) annual conference on technological advancements in developing countries, Columbia University, New York
Bharath A, Madhvanath S (2007) Hidden Markov models for online handwritten tamil word recognition. Ninth Int Conf Doc Anal & Recognit, 1:506–510
Bharath A, Madhvanath S (2012) Hmm-based lexicon-driven and lexicon-free word recognition for online handwritten indic scripts. IEEE Trans Pattern Anal & Mach Intell, 34:670–682
Bhattacharya N, Pal U, Roy PP (2017) Stroke-order normalization for online bangla handwriting recognition. \(14^{th}\) IAPR Int Conf Doc Anal & Recognit, 1:206–211
Bhunia AK, Das A, Roy PP et al (2015) A comparative study of features for handwritten bangla text recognition. In: 2015 13th International Conference on Document Analysis and Recognition (ICDAR), pp 636–640, https://doi.org/10.1109/ICDAR.2015.7333839
Bouslama F, Amin A (1998) Pen-based recognition system of arabic character utilizing structural and fuzzy techniques. Second international conference on knowledge-based intelligent electronic systems 3:76–85
Budsayaplakorn R, Asdornwised W, Jitapunkul S (2003) On-line thai handwritten character recognition using hidden markov model and fuzzy logic. IEEE \(^{th}\) workshop on neural networks for signal processing 13:537–546
Chakraborty B, Mukherjee PS, Bhattacharya U (2016) Bangla online handwriting recognition using recurrent neural network architecture. In: Proceedings of the tenth Indian conference on computer vision. Graphics and image processing, 1—8
Chowdhury K, Alam L, Sarmin S et al (2015) A fuzzy features based online handwritten bangla word recognition framework. \(18^{th}\) International Conference on Computer and Information Technology pp 484–489
Chowdhury S, Garai U, Chattopadhyay T (2011) A weighted finite-state transducer (wfst)-based language model for online indic script handwriting recognition. In: International Conference on Document Analysis and Recognition, pp 599—602
Das N, Sarkar R, Basu S et al (2015) Handwritten bangla character recognition using a soft computing paradigm embedded in two pass approach. Pattern Recognit, 48:54–2071
Eberhard DM, Gary SF, Charles FD (2020) Ethnologue: Languages of the world. In: Dallas. Texas: Sil International
El-Wakil MS, Shoukry AA (1989) On-line recognition of handwritten isolated arabic characters. Pattern Recognit, 22:97–105
Fink GA, Vajda S, Bhattacharya U et al (2010) Online bangla word recognition using sub-stroke level features and hidden Markov models. \(12^{th}\) International Conference on Frontiers in Handwriting Recognition pp 393–398
Frinken V, Bhattacharya N, Pal U (2014) Design of unsupervised feature extraction system for on-line bangla handwriting recognition. \(11^{th}\) IAPR International Workshop on Document Analysis Systems pp 355–359
Ghods V, Kabir E, Razzazi F (2013) Effect of delayed strokes on the recognition of online farsi handwriting. Pattern Recogn Lett, 34(5):486–491
Ghosh R, Roy PP (2016) Comparison of zone-features for online bengali and devanagari word recognition using hmm. \(15^{th}\) International Conference on Frontiers in Handwriting Recognition pp 435–440
Ghosh R, Vamshi C, Kumar P (2019) Rnn based online handwritten word recognition in devanagari and bengali scripts using horizontal zoning. Pattern Recognit, 92:203–218
Halavati R, Shouraki SB (2007) Recognition of persian online handwriting using elastic fuzzy pattern recognition. Int J Pattern Recognit & Artif Intell 21(3):491–513
Hochreiter S, Schmidhuber J (1997) Long short-term memory. Neural Comput, 9:1735–1780
Izadi S, Haji M, Suen CY (2008) A new segmentation algorithm for online handwritten word recognition in persian script. In: Proceedings of the eleventh international conference on frontiers in handwriting recognition, pp 598—603
Jäger S, Liu CL, Nakagawa M (2003) The state of the art in japanese online handwriting recognition compared to techniques in western handwriting recognition. Doc Anal & Recognit, 6(2):75–88
Karnchanapusakij C, Suwannakat P, Rakprasertsuk W, et al (2009) Online handwriting thai character recognition. In: Sixth international conference on computer graphics imaging and visualization, pp 323—328
Liu CL, Zhou XD (2006) Online Japanese character recognition using trajectory-based normalization and direction feature extraction. Tenth international workshop on frontiers in handwriting recognition, Suvisoft
Liu CL, Jaeger S, Nakagawa M (2004) Online recognition of chinese characters: the state-of-the-art. IEEE Trans Pattern Anal & Mach Intell, 26(2):198–213
Liu CL, Yin F, Wang DH et al (2013) Online and offline handwritten chinese character recognition: benchmarking on new databases. Pattern Recognit, 46(1):155–162
Matsumoto K, Fukushima T, Nakagawa M (2001) Collection and analysis of on-line handwritten japanese character patterns. In: Sixth International conference on document analysis and recognition, pp 496—500
Mohiuddin S, Bhattacharya U, Parui SK (2011) Unconstrained bangla online handwriting recognition based on mlp and svm. In: Proceedings of the 2011 Joint Workshop on Multilingual OCR and Analytics for Noisy Unstructured Text Data, pp 1–6
Mukherjee PS, Bhattacharya U, Parui SK et al (2017) A hybrid model for end-to-end online handwriting recognition. \(14^{th}\) IAPR International Conference on Document Analysis and Recognition 1:658–663
Obaidullah SM, Halder C, Santosh K et al (2018) Phdindic_11: Page-level handwritten document image dataset of 11 official indic scripts for script identification. Multimed Tools & Appl, 77(2):1643–1678. https://doi.org/10.1007/s11042-017-4373-y
Parui SK, Guin K, Bhattacharya U et al (2008) Online handwritten bangla character recognition using hmm. \(19^{th}\) International conference on pattern recognition pp 1–4
Plamondon R, Srihari SN (2000) Online and off-line handwriting recognition: a comprehensive survey. IEEE Trans Pattern Anal & Mach Intell, 22(1):63–84
Razavi S, Kabir E (2004) A data base for online persian handwritten recognition. \(6^{th}\) Conference on intelligent systems pp 859–863
Sanguansat P, Asdornwised W, Jitapunkul S (2004) Online thai handwritten character recognition using hidden Markov models and support vector machines. IEEE Int Symp Commun & Inf Tech, 1:492–497
Sen S, Bhattacharyya A, Mitra M et al (2020) Online bangla handwritten word recognition using hmm and language model. Neural Comput Appl, 32:9939–9951
Singh H, Sharma RK, Singh VP (2020) Online handwriting recognition systems for indic and non-indic scripts: a review. Artif Intell Rev, pp 1573–7462
Srimony A, Chowdhuri SD, Bhattacharya U et al (2014) Holistic recognition of online handwritten words based on an ensemble of svm classifiers. \(11^{th}\) IAPR International Workshop on Document Analysis Systems pp 86–90
Takahashi K, Yasuda H, Matsumoto T (1997) A fast hmm algorithm for on-line handwritten character recognition. Proc fourth Int Conf Doc Anal & Recognit, 1:369–375
Tappert CC, Suen CY, Wakahara T (1990) The state of the art in online handwriting recognition. IEEE Trans Pattern Anal & Mach Intell, 12(8):787–808
Veltman SR, Prasad R (1994) Hidden markov models applied to on-line handwritten isolated character recognition. IEEE Trans Image Proc, 3(3):314–318
Zhou XD, Yu JL, Liu CL et al (2007) Online handwritten japanese character string recognition incorporating geometric context. In: Ninth International conference on document analysis and recognition, pp 48—52
Acknowledgements
One of the authors would like to thank SERB, DST for financial support in the form of a project
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflicts 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.
Rights and permissions
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
Bhattacharyya , A., Chatterjee, S., Sen, S. et al. BWordDeepNet: a novel deep learning architecture for the recognition of online handwritten Bangla words. Multimed Tools Appl 83, 45071–45093 (2024). https://doi.org/10.1007/s11042-023-16709-8
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-023-16709-8