Advancing Optical Character Recognition for Handwritten Text: Enhancing Efficiency and Streamlining Document Management
Benedict Vinusha. V, V Indhuja., Varshitha Reddy M, Nagalla Nikhitha, Naga Venkata Siva Reddy · 2023
This research paper explores the challenge of converting handwritten text into editable digital text, in both native and English languages. We propose optical character recognition (OCR) as a solution, leveraging algorithms and machine learning models to recognize and convert text into various digital formats. The paper highlights the significance of OCR in digitizing paper documents and streamlining document management across diverse industries. Specifically, the study focuses on developing OCR software capable of identifying regional languages, providing a comprehensive assessment of its precision and effectiveness. The findings demonstrate how OCR technology can significantly improve efficiency and cost-effectiveness for businesses dealing with large volumes of paper documents. Our approach involves using image inputs, analyzing them with Keras and OpenCV, and implementing Convolutional Neural Network (CNN) techniques and Natural Language Processing (NLP) libraries.