Authors : Dr. T. Nirmal Raj; T. Ramya

Volume/Issue : Volume 11 - 2026, Issue 5 - May

Google Scholar : https://tinyurl.com/4tv9nfvj

Scribd : https://tinyurl.com/bd5yyd26

DOI : https://doi.org/10.38124/ijisrt/26May1370

Note : A published paper may take 4-5 working days from the publication date to appear in PlumX Metrics, Semantic Scholar, and ResearchGate.

Abstract : With the rapid advancement of digital technologies and the increasing demand for personalized digital content, the need to convert handwritten text into reusable and scalable digital formats has become highly significant. In particular, regional languages like Tamil, which possess rich script structures and cultural importance, require efficient digitization techniques to preserve and promote their usage in modern digital environments. This project proposes a comprehensive system designed to convert handwritten Tamil text into a fully functional digital font. The system allows users to write Tamil text manually on paper, scan or capture the document as an image, and upload it into the application. The uploaded handwritten document undergoes a series of image preprocessing steps, including noise removal, grayscale conversion, binarization, normalization, and segmentation, to enhance the quality and prepare the data for accurate recognition. Following preprocessing, the system employs character recognition techniques to identify individual Tamil characters. Feature extraction methods are applied to capture the unique structural patterns of each handwritten character. These features are then analyzed using machine learning or pattern recognition algorithms to accurately classify and map each character to its corresponding Tamil Unicode representation. Once the characters are recognized, the system proceeds to generate scalable digital glyphs. Each glyph is carefully designed to retain the stylistic characteristics of the user’s handwriting, ensuring personalization. These glyphs are then assembled into a complete font file (such as TTF or OTF format), which can be installed and used across various applications. The system also provides an interactive user interface that enables users to preview the generated font in real-time. Users can test the font with sample text, make adjustments if necessary, and download the finalized font for use in documents, graphicdesign, publishing, and other digital platforms.

References :

1. Gonzalez, R. C., & Woods, R. E. Digital Image Processing, 4th Edition, Pearson Education, 2018.
2. Pratt, W. K. Digital Image Processing: PIKS Scientific Inside, Wiley-Interscience, 2007.
3. Bishop, C. M.Pattern Recognition and Machine Learning, Springer, 2006.
4. Goodfellow, I., Bengio, Y., & Courville, A.
5. Deep Learning, MIT Press, 2016. Pal, U., & Chaudhuri, B. B.“Indian Script Character Recognition: A Survey,” Pattern Recognition, Elsevier, 2004.
6. Arora, S., Bhattacharjee, D., Nasipuri, M., Basu, D. K., & Kundu, M. “Performance Comparison of SVM and ANN for Handwritten Devanagari Character Recognition,” IJCSI, 2010.
7. Jawahar, C. V., Kumar, M. N., & Ravindran, B. “A Bilingual OCR System for Hindi and Tamil,” ICDAR Conference Proceedings, 2007.
8. LeCun, Y., Bottou, L., Bengio, Y., & Haffner, P. “Gradient-Based Learning Applied to Document Recognition,” Proceedings of the IEEE, 1998.