Composite Artificial Intelligence
by T. S. Arun Samuel, L. Jerart Julus, P. Kanimozhi, T. Ananth Kumar, S. Balamurugan
3A Composite Artificial Intelligence Framework for Enhanced and Intelligent Word Recognition of Handwritten Hindi
R. S. Rampriya1*, Sabarinathan2, SahayaBeni Prathiba3, C. Renit4, R. Arumuga Arun5 and S. Bhuvana1
1School of Computer Science Engineering and Information Systems, Vellore Institute of Technology, Vellore, Tamilnadu, India
2Department of Artificial Intelligence, Couger Inc., Tokyo, Japan
3Centre for Cyber Physical Systems, School of Computer Science and Engineering, Vellore Institute of Technology, Chennai, India
4Department of ECE, St. Xaviers Catholic College of Engineering, Kanyaumari, Tamilnadu, India
5School of Computer Science Engineering, Vellore Institute of Technology, Vellore, Tamilnadu, India
Abstract
The Hindi Handwritten Intelligent Character Recognition (HICR) system digitizes handwritten Hindi text, enhancing accessibility, preserving cultural heritage, and improving efficiency in education and administration. It supports multilingual communication and modernizes various sectors by integrating advanced technology with linguistic diversity. This study uses the combination of two AI models to provide a novel method for handwritten Hindi word recognition: HindRoBERTa (A Hindi Robustly Optimized BERT Pretraining Approach) decoder and the Data-efficient Image Transformer (DeiT) encoder, within a Text Recognition Transformer (TROCR) framework. A vision transformer (ViT) model which is acted as encoder, obtaining the characteristics from handwritten Hindi ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access