IITBBS-OCR-Dataset

Odia is a classical and popular language in the Indian subcontinent used by more than 50 million people. In spite of its rich history, popularity and usefulness, not much research efforts have been made to achieve high level accuracy in case of Odia OCR. New handwritten alphanumeric character and numeral datasets for Odia are created by our research group@iitbbs and reported here in order to address the paucity of benchmark Odia datasets. We hope that these publicly available Odia numeral and character datasets will help in creating high accuracy Odia character recognition systems. Please also see the relevant reseach works reported by our research group@iitbbs and cite them appropriately in your novel findings or publications.

For more clarification, please contact: Dr. Niladri B. Puhan, Associate Professor, IIT Bhubaneswar, India: nbpuhan@iitbbs.ac.in