Handwritten Alphanumeric Character Recognition using Jetson Nano

Author : Aruna Rao S. L, Chatragadda Bharani, Jaina Sri Laxmi, Soppadandi Spoorthi, Kummari Tejaswini

DOI : 10.5072/FK26H4PV9J.2023.05.08.001

Handwritten Alphanumeric Character Recognition is one of the significant areas of exploration and development with a streaming number of possibilities that could be attained. The applications of alphanumeric character recognition include postal correspondence sorting, bank cheque processing, form data entry, etc. In most of the applications, major challenge lies in copying the contents from original file where the content may be in a noneditable format. The heart of the project lies within the ability to develop an efficient algorithm that can recognize handwritten alphanumeric characters which are submitted by users, which may vary in their font styles and font sizes. In order to implement this, we used EMNIST Balanced Character Dataset to train the Machine Learning model using Deep Learning. Flask is used for API and user interface. The goal is to deploy whole system on Jetson Nano Developer Kit with an optimal solution and the best accuracy which is 87%

Full Text Attachment