Review of Character Recognition Techniques for MODI Script

Abstract

<h2>Abstract</h2>\n<p><strong>Objectives:</strong> The main objective of this study is to review and compare the various methods used for Modi script recognition.<strong> Methods:</strong> The author has chosen various methods from 2010 to 2022 that are used to process MODI Script. The distinct methods employed for feature extraction and classification are compared for various datasets. A discussion on the significance of the selection of correct feature extraction and classification techniques and the comments on the methods suited to specific applications is provided. <strong>Findings:</strong> Currently, there are very few MODI translators. In contrast, millions of historical documents written in MODI remain unexplored.<strong> Novelty:</strong> The Convolutional Neural Networks (CNNs) has been used successfully for recognizing MODI Script characters. In the present study the author finds that, compared to all techniques, CNN provides maximum accuracy of 99.78%. Hence, CNN is the best character recognition technique for the MODI script.</p>\n<p><strong>Keywords:</strong> Distance Classifier; Feature Extraction; and Classification; Handwritten Characters; MODI Lipi; Character Recognition</p>