Leveraging the Swin Transformer for Enhanced Handwritten Urdu Character and Digit Recognition
Keywords:
Swin Transformer, character recognition, handwritten Urdu characters, digit recognition, CNN, LSTMAbstract
The Swin Transformer model is used to identify handwritten Urdu characters and digits. The paper also focuses on the challenges associated with the script of Urdu, a language that uses an Arabic-based alphabet. It also explores its ability to read different character types and scripts. This is achieved by training on a large dataset and then testing it for character recognition ability. “MANUU: Handwritten Urdu OCR Dataset” was employed in order to measure the performance of the model in terms of accuracy, precision, recall and F1-score. However, The Swin Transformer outperforms existing CNN and LSTM classifiers in accordance with the higher levels of accuracy (97%), F1-score (89%), recall (91%) as well as precision (83%). With its benefits over traditional classifiers, these results further demonstrate how efficient Swin Transformer performs better than current CNN or LSTM classifiers for achieving higher levels of accuracy (97%), F1-score (89%), recall (91%) as well as precision (83%). So, this research highlights how valuable the Swin transformer is when it comes to recognizing correct Urdu letters besides comparing it with other traditional classifiers that fail in this regard. In addition, there are certain things within this study that show how effective can be Swin transformer while solving problems which are particular to some languages pointing at his importance within bigger framework of language processing.
Downloads
References
M. A. Kilvisharam Oziuddeen, S. Poruran, and M. Y. Caffiyar, “A novel deep convolutional neural network architecture based on transfer learning for handwritten urdu character recognition,” Teh. Vjesn., vol. 27, no. 4, pp. 1160–1165, 2020, doi: 10.17559/TV-20190319095323.
Q. A. Safdar, K. U. Khan, and L. Peng, “A novel similar character discrimination method for online handwritten Urdu character recognition in half forms,” Sci. Iran., vol. 29, no. 5, pp. 2419–2436, 2022.
M. N. AlJarrah, M. Z. Mo’ath, and R. Duwairi, “Arabic handwritten characters recognition using convolutional neural network,” in 2021 12th International Conference on Information and Communication Systems (ICICS), IEEE, 2021, pp. 182–188.
H. Zargar, R. Almahasneh, and L. T. Kóczy, “Automatic recognition of handwritten Urdu characters,” Comput. Intell. Math. Tackling Complex Probl. 3, pp. 165–175, 2022.
I. Uddin et al., “Benchmark pashto handwritten character dataset and pashto object character recognition (OCR) using deep neural network with rule activation function,” Complexity, vol. 2021, pp. 1–16, 2021.
S. Khan and S. Nazir, “Deep learning based Pashto characters recognition: LSTM-based handwritten Pashto characters recognition system,” Proc. Pakistan Acad. Sci. A. Phys. Comput. Sci., vol. 58, no. 3, pp. 49–58, 2021.
R. Guha, N. Das, M. Kundu, M. Nasipuri, and K. C. Santosh, “Devnet: an efficient cnn architecture for handwritten devanagari character recognition,” Int. J. Pattern Recognit. Artif. Intell., vol. 34, no. 12, p. 2052009, 2020.
W. Jiang, “Evaluation of deep learning models for Urdu handwritten characters recognition,” in Journal of Physics: Conference Series, IOP Publishing, 2020, p. 12016.
V. K. Chauhan, S. Singh, and A. Sharma, “HCR-Net: A deep learning based script independent handwritten character recognition network,” arXiv Prepr. arXiv2108.06663, 2021.
R. M. Ahmed et al., “Kurdish Handwritten character recognition using deep learning techniques,” Gene Expr. Patterns, vol. 46, p. 119278, 2022.
F. Anwar, “Online urdu handwritten text recognition for mobile devices using intelligent techniques.” International Islamic University, Islamabad, 2019.
S. S. R. Rizvi, A. Sagheer, K. Adnan, and A. Muhammad, “Optical character recognition system for Nastalique Urdu-like script languages using supervised learning,” Int. J. Pattern Recognit. Artif. Intell., vol. 33, no. 10, p. 1953004, 2019.
K. O. Aarif and P. Sivakumar, “Multi-Domain Deep Convolutional Neural Network for Ancient Urdu Text Recognition System.,” Intell. Autom. Soft Comput., vol. 33, no. 1, 2022.
H. Ali, A. Ullah, T. Iqbal, and S. Khattak, “Pioneer dataset and automatic recognition of Urdu handwritten characters using a deep autoencoder and convolutional neural network,” SN Appl. Sci., vol. 2, pp. 1–12, 2020.
M. M. Misgar, F. Mushtaq, S. S. Khurana, and M. Kumar, “Recognition of offline handwritten Urdu characters using RNN and LSTM models,” Multimed. Tools Appl., vol. 82, no. 2, pp. 2053–2076, 2023.
M. S. Amin, S. M. Yasir, and H. Ahn, “Recognition of pashto handwritten characters based on deep learning,” Sensors, vol. 20, no. 20, p. 5884, 2020.
T. Kattenborn, J. Leitloff, F. Schiefer, and S. Hinz, “Review on Convolutional Neural Networks (CNN) in vegetation remote sensing,” ISPRS J. Photogramm. Remote Sens., vol. 173, no. July 2020, pp. 24–49, 2021, doi: 10.1016/j.isprsjprs.2020.12.010.
L. Alzubaidi et al., Review of deep learning: concepts, CNN architectures, challenges, applications, future directions, vol. 8, no. 1. Springer International Publishing, 2021. doi: 10.1186/s40537-021-00444-8.
H. Salman, J. Grover, and T. Shankar, “Hierarchical Reinforcement Learning for Sequencing Behaviors,” vol. 2733, no. March, pp. 2709–2733, 2018, doi: 10.1162/NECO.
K. Smagulova and A. P. James, “A survey on LSTM memristive neural network architectures and applications,” Eur. Phys. J. Spec. Top., vol. 228, no. 10, pp. 2313–2324, 2019, doi: 10.1140/epjst/e2019-900046-x.
Sanjay Kumar Suman, Dhananjay Kumar and L. Bhagyalakshmi, “SINR pricing in non- cooperative power control game for wireless ad hoc network”, KSII Transactions on Internet and Information Systems, KSII TIIS, vol. 8, no. 7, pp. 2281-2301, 2014. https://dio.org/10.3837/tiis.2014.07.005
L. Bhagyalakshmi, Sanjay Kumar Suman, Sujeetha Devi, “Joint Routing and Resource Allocation for Cluster Based Isolated Nodes in Cognitive Radio Wireless Sensor Networks”, Wireless Personal Communication, Springer, vol. 114, issue 4, pp. 3477- 3488, Oct. 2020. https://doi.org/10.1007/s11277-020-07543-4
K. Mahalakshmi, K. Kousalya, Himanshu Shekhar, Aby K. Thomas, L. Bhagyalakshmi, Sanjay Kumar Suman et. al., “Public Auditing Scheme for Integrity Verification in Distributed Cloud Storage System”, Scientific Programming, Hindawi, vol. 2021, Article ID 8533995, Dec. 2021. https://doi.org/10.1155/2021/8533995
Sanjay Kumar Suman et. al., Detection and prediction of HMS from drinking water by analysing the adsorbents from residuals using deep learning, Hindawi (SAGE Journal) Adsorption Science & Technology, vol. 2022, Article id 3265366, March 2022. https://doi.org/10.1155/2022/3265366.
Bhagyalakshmi and K. Murugan, “Avoiding Energy Holes Problem using Load Balancing Approach in Wireless Sensor Network”, KSII Transaction on Internet and Information Systems, vol. 8, no. 5, pp. 1618- 1637, 2014. https://dio.org/10.3837/tiis.2014.05.007.
Satyanand Singh, Sajai Vir Singh, Dinesh Yadav, Sanjay Kumar Suman, Bhagyalakshmi Lakshminarayanan, Ghanshyam Singh, “Discrete interferences optimum beamformer in correlated signal and interfering noise, International Journal of Electrical and Computer Engineering, vol 12, no. 2, pp. 1732-1743, April 2022. http://doi.org/10.11591/ijece.v12i2.pp1732-1743:
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
All papers should be submitted electronically. All submitted manuscripts must be original work that is not under submission at another journal or under consideration for publication in another form, such as a monograph or chapter of a book. Authors of submitted papers are obligated not to submit their paper for publication elsewhere until an editorial decision is rendered on their submission. Further, authors of accepted papers are prohibited from publishing the results in other publications that appear before the paper is published in the Journal unless they receive approval for doing so from the Editor-In-Chief.
IJISAE open access articles are licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. This license lets the audience to give appropriate credit, provide a link to the license, and indicate if changes were made and if they remix, transform, or build upon the material, they must distribute contributions under the same license as the original.
