Odia Handwritten Character Recognition Based on Convolutional Neural Network

EasyChair Preprint 7713

Abstract

Odia is an eastern Indo-Aryan language spoken by 44 million people especially in Odisha, India .Some character are made up of more than one connected symbols. Handwritten Character Recognition (HCR) plays an important role in Optical Character Recognition (OCR) and Pattern Recognition (PR), as it has a good number of applications in various fields .HCR contributes extremely to the growth of automation and are applicable in the areas of bank cheque, medical prescriptions, tax returns etc. Handwritten characters are much more difficult to recognize than the printed characters due to different in writing styles for different people. We can use the character recognition technology that scans the different types of document that can be either handwritten or printed. This paper explores and analyzes the use of Deep learning techniques such as the Convolutional Neural Networks for the recognition of Odia characters. Inspired by the structure of the brain, CNNs classify characters by making use of neurons linked in various layers so as to achieve maximum efficiency. In this paper, 6 layers of neurons were used for the purpose of classifying Odia characters. An accuracy of 95.6% was achieved in this approach. Once recognized, the handwritten Odia characters can be easily translated into English or any other languages.

Keyphrases: Convolutional Neural Networks, Odia Script, deep learning, image processing

@booklet{EasyChair:7713,
  author    = {Rabinarayan Panda and Sachikanta Dash and Sasmita Padhy},
  title     = {Odia Handwritten Character Recognition Based on Convolutional Neural Network},
  howpublished = {EasyChair Preprint 7713},
  year      = {EasyChair, 2022}}