A Novel Artificial Intelligence Approach to Optical Character Recognition of Conjunct Gujarati Script
DOI:
https://doi.org/10.32628/IJSRST2513114Keywords:
Gujarati OCR, conjunct characters, machine learning, deep learning, natural language processing, Indic script recognitionAbstract
This paper surveys recent advances in optical char- acter recognition (OCR) for the Gujarati script, with a focus on complex conjunct characters. Gujarati is an Indo-Aryan script spoken by ∼62 million people [1], [2], yet its OCR remains challenging due to intricate glyph shapes and extensive consonant clusters [3], [4]. We review how machine learning (ML), deep learning (DL), and NLP techniques have been applied to segment and recognize Gujarati text, especially conjunct lig- atures. Notable studies from 2012–2025 are examined, including ANN and CNN-based classifiers that achieve high accuracy on isolated conjuncts [3], [5]. Finally, ongoing challenges (data scarcity, variability of handwriting and fonts) and outline future directions such as transformer models and language-model integration for Gujarati OCR.
📊 Article Downloads
References
B. Panchal, A. Shah, “A Survey on Gujarati NLP Research Work,” SSRN, 2025. DOI: https://doi.org/10.15649/2346030X.4445
M. Patel, “Identification of Offline Gujarati Handwritten Conjunct Characters,” IRJET, 2021.
B. Patel, “Identification of Typewritten and Handwritten Conjunct Gu- jarati Characters Using ANN,” IJAPR, 2022. DOI: https://doi.org/10.1504/IJAPR.2022.122267
C. Patel, A. Desai, “Extraction of Characters and Modifiers from Handwritten Gujarati Words,” IJCA, 2013. DOI: https://doi.org/10.5120/12719-9541
M. Parikh, A. Desai, “Recognition of Handwritten Gujarati Conjuncts Using CNN Architectures,” ICACDS, 2022.
M. Parikh, A. Desai, “A Novel ConvNet Architecture for Gujarati Conjuncts,” LNNS, Springer, 2025.
Y. Zala et al., “Handwritten Gujarati Character Recognition Using ML and DL,” ICAMIDA, 2023. DOI: https://doi.org/10.2991/978-94-6463-136-4_76
A. Bhuva, D. Mishra, “Gujarati OCR Using Efficient Text Feature Extraction,” Informatica, 2025.
R. Kundal, B. Parekh, “Deep Learning for Handwritten Gujarati Script,” Revista Electronica de Veterinaria, 2024.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 International Journal of Scientific Research in Science and Technology
This work is licensed under a Creative Commons Attribution 4.0 International License.
https://creativecommons.org/licenses/by/4.0
