Proceedings of the 2025 International Conference on Advanced Research in Electronics and Communication Systems (ICARECS-2025)

<Previous Article In Volume
Next Article In Volume>

Handwritten Tamil Text Recognition Using Vision Transformer: A Novel Approach

Authors
A. Aarthi Abiramy1, *, P. Ramkumar1, S. Gurusriram1, C. Ajitha1
1Department of Information Technology, National Engineering College, Kovilpatti, India
*Corresponding author. Email: 2115024@nec.edu.in
Corresponding Author
A. Aarthi Abiramy
Available Online 30 June 2025.
DOI
10.2991/978-94-6463-754-0_34How to use a DOI?
Keywords
Self Attention; Vision Transformer (ViT); Encoder-Decoder Architecture; Convolutional Neural Network (CNN); Image Patching
Abstract

The field of handwritten character recognition has seen remarkable strides with the advent of deep learning techniques. Whereas Convolutional Neural Networks (CNNs) have proven to be extremely effective in handwritten text recognition for most languages, newer breakthroughs in Vision Transformers (ViT) present new avenues for enhancing accuracy and efficiency in such recognition tasks. This contribution considers the application of ViT for handwritten Tamil script recognition, characterized by complex characters and complex diacritical marks. The proposed model takes advantage of the self-attention mechanism characteristic of transformers to capture nuanced patterns of the Tamil script. Experimental outcomes suggest that ViT is more accurate than standard CNN-based models. We applied ViT for enhancing the accuracy of recognition as well as the robustness of the model. Particularly, this paper discusses special difficulties in handwritten Tamil text recognition.

Copyright
© 2025 The Author(s)
Open Access
Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

<Previous Article In Volume
Next Article In Volume>
Volume Title
Proceedings of the 2025 International Conference on Advanced Research in Electronics and Communication Systems (ICARECS-2025)
Series
Atlantis Highlights in Engineering
Publication Date
30 June 2025
ISBN
978-94-6463-754-0
ISSN
2589-4943
DOI
10.2991/978-94-6463-754-0_34How to use a DOI?
Copyright
© 2025 The Author(s)
Open Access
Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Cite this article

risenwbib
TY  - CONF
AU  - A. Aarthi Abiramy
AU  - P. Ramkumar
AU  - S. Gurusriram
AU  - C. Ajitha
PY  - 2025
DA  - 2025/06/30
TI  - Handwritten Tamil Text Recognition Using Vision Transformer: A Novel Approach
BT  - Proceedings of the 2025 International Conference on Advanced Research in Electronics and Communication Systems (ICARECS-2025)
PB  - Atlantis Press
SP  - 390
EP  - 400
SN  - 2589-4943
UR  - https://doi.org/10.2991/978-94-6463-754-0_34
DO  - 10.2991/978-94-6463-754-0_34
ID  - Abiramy2025
ER  -