Abstract
Due to digitalization in everyday life, the need for automatically recognizing handwritten digits is increasing. Handwritten digit recognition is essential for countless applications in various industries. Bengali ranks the fifth largest dialect in the world, with 265 million speakers (native and non-native combined), occupying 4% of the world population. Due to the complexity of Bengali writing in terms of variety in shape, size, and writing style, researchers did not get better accuracy using supervised machine learning algorithms to date. Moreover, fewer studies have been done on Bangla handwritten digit recognition (BHwDR). In this paper, a novel convolutional neural network (CNN)-based pre-trained handwritten digit recognition model has been proposed, which includes ResNet-50, Inceptionv3, and EfficientNetB0 on the NumtaDB dataset of 17 thousand instances with ten classes. The result outperformed the performance of other models to date with 97% accuracy in the 10-digit classes. Furthermore, we have evaluated the result of our model with other research studies while suggesting future studies.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Eberhard, D.M., Simons, G.F., Fennig, C.D. (eds.): Ethnologue: Languages of the World, 25 edn. SIL International, Dallas, Texas (2022). Online version: http://www.ethnologue.com
Zanwar, S.R., Shinde, U.B., Narote, A.S., Narote, S.P.: Handwritten English character recognition using swarm intelligence and neural network. In: Springer, vol. 1148, pp. 93–102 (2020)
Gan, J., Wang, W., Lu, K.: Compressing the CNN architecture for in-air handwritten Chinese character recognition. ScienceDirect 129, 190–197 (2020)
Granell, E., Chammas, E., Likforman-Sulem, L., Martínez-Hinarejos, C.D., Mokbel, C., Cîrstea, B.I.: Transcription of Spanish historical handwritten documents with deep neural networks. J. Imaging 4(1), 15 (2018)
Boufenar, C., Batouche, M., Schoenauer, M.: An artificial immune system for offline isolated handwritten arabic character. Springer 9, 25–42 (2018)
Mukhoti, J., Dutta, S., Sarkar, R.: Handwritten digit classification in Bangla and Hindi using deep learning. Appl. Artif. Intell. 34, 1–26 (2020)
Nanehkaran, Y.A., Zhang, D., Salimi, S., Che, J.: Analysis and comparison of machine learning classifiers and deep neural networks techniques for recognition of Farsi handwritten digits. Springer 77, 3193–3222 (2021)
Papantoniou, K., Tzitzikas, Y.: NLP for the Greek Language: a brief survey. In: Proceedings of the 11th Hellenic Conference on Artificial Intelligence. ACM, pp. 101–109 (2020)
Bhowmik, T.K., Bhattacharya, U., Parui, S.K.: Recognition of Bangla handwritten characters using an MLP classifier based on stroke features. In: Springer, pp. 814–819 (2004)
Basu, S., Das, N., Sarkar, R., Kundu, M., Nasipuri, M., Basu, D.K.: Handwritten Bangla alphabet recognition using an MLP based Classifier. arxiv (2012)
Hamdan, Y.B., Sathesh, A.: Deep learning based handwriting recognition with adversarial feature deformation and regularization. J. Innovative Image Process. 3(4), 367 (2021)
Vijayakumar, T.: Synthesis of palm print in feature fusion techniques for multimodal biometric recognition system online signature. J. Innovative Image Process. (JIIP) 3(02), 131–143 (2021)
Gayathri, R., Babitha Lincy, R.: Transfer learning based handwritten character recognition of Tamil script using inception-V3 model. 6091–6102
Basri, R., et al.: Bangla handwritten digit recognition using deep convolutional neural network. In: Proceedings of the International Conference on Computing Advancements (2020)
Tallapragada, V.V., et al.: Greek handwritten character recognition using Inception V3. In: Smart Systems: Innovations in Computing, pp. 247–257. Springer, Singapore (2022)
Chatterjee, S., Dutta, R.K., Ganguly, D., Chatterjee, K., Roy, S.: Bengali handwritten character classification using transfer learning on deep convolutional network. In: International Conference on Intelligent Human Computer Interaction, pp. 138–148. Springer, Cham (2019)
Mondal, S., Mahfuz, N.: Convolutional neural networks based Bengali handwritten character recognition. In: Springer, vol. 325, pp. 718–729 (2020)
Dalal, S., Dastoor, S., Zaveri, K., Choksi, V., Shah, K.: Performance analysis of Gujarati script recognition using multiclass and multilabel classification. In: Tuba, M., Akashe, S., Joshi, A. (eds.) ICT Systems and Sustainability. Lecture Notes in Networks and Systems, vol. 321. Springer, Singapore (2022). https://doi.org/10.1007/978-981-16-5987-4_72
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
Krizhevsky, A., Sutskever, I., Hinton, G.E.: Imagenet classification with deep convolutional neural networks. Commun. ACM 60(6), 84–90 (2017)
Alam, S., Reasat, T., Doha, R.M., Humayun, A.I.: Numtadb-assembled Bengali handwritten digits (2018). arXiv preprint arXiv:1806.02452
Author information
Authors and Affiliations
Contributions
Islam M significantly contributed to the conceptual parts of the paper’s contribution to the knowledge. Then, Islam M. and Shuvo SA devised the methodology for the paper and conducted the experiment. Shaikh MM and Sourav MSU contributed to the section on literature review. Nipun MS and Nayeem J have examined the experiment's outcomes and contributed to the section’s design and writing. Abstract, introduction, and conclusion were all written by Haque Z. Sulaiman RB oversaw the entire project and contributed to every aspect of this study article. Shuvo SA has contributed to the administration of group collaboration on a global scale. Sourav MSU assisted in finalizing the manuscript. Ameer K contributed to the project's finances by covering conference registration fees and other costs. All authors examined the findings and approved the final manuscript version.
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Islam, M. et al. (2023). Efficient Approach to Using CNN-Based Pre-trained Models in Bangla Handwritten Digit Recognition. In: Smys, S., Tavares, J.M.R.S., Shi, F. (eds) Computational Vision and Bio-Inspired Computing. Advances in Intelligent Systems and Computing, vol 1439. Springer, Singapore. https://doi.org/10.1007/978-981-19-9819-5_50
Download citation
DOI: https://doi.org/10.1007/978-981-19-9819-5_50
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-19-9818-8
Online ISBN: 978-981-19-9819-5
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)