Optical/Handwritten Data Recognition

Abstract

For many years, optical character recognition (OCR) has been a hot topic. It's the process of breaking down a document image into its individual characters. Despite decades of intensive research, producing OCR with human-like skills is still a work in progress. The industries have long used localization and recognition of written characters for a number of applications. When working in a sterile environment, such as scanners or static settings, most optical character recognition (OCR) systems function well. The demand for OCR in the natural environment is increasing as technology advances. The most notable example is the recognition and localization of text outside, when the settings are far from ideal for machine vision applications. The main goal of this project is to create an Web based application that can recognise text content in JPEG photographs, PDF files, or photos shot in real time and convert them into an editable content document for several languages. For the same reason, this study focuses on the translation of written materials into speech. Due to the difficulty of the task, industrial and university researchers have turned their attention to Optical Character Recognition. The number of academic laboratories and businesses engaged in character recognition research has risen considerably in recent years. Optical character recognition (OCR) allows for a variety of automation applications. The goal of this research is to discover and recognise words in natural photographs. The targeted challenge is substantially more difficult than interpreting text from scanned documents. The goal of this study is to summarise the existing research in the topic of OCR. It gives an overview of many aspects of OCR and explores relevant solutions for fixing OCR concerns. The project had an accuracy rate of more than 80% in character recognition and more than 60% in handwritten character recognition. This document describes the project's stages of development, main hurdles, and some of the project's most noteworthy findings.