Optical Character Recognition Using Tesseract

Abstract

Abstract: Optical Character Recognition (OCR) is a process or technology in which text within a digital image is recognized. With rapid pace of technology, people want quicker, handy and reliable tools, which can fulfil their daily needs. With this moto we had gone forward and analyzed the existing tools and made up this Android App, which provides seamless experience (No ads and easy-to-use), and great accuracy. The main objective of this project is to allow automatic extraction of the information that a user wants from the paper document and using it wherever it is needed. In this project, OCR uses Tesseract as an engine to display the text to the user and uses a Deep learning model to classify the letters and display them to the user. It adds a new neural network (LSTM) based OCR engine which is focused on line recognition but also still supports the legacy Tesseract OCR engine which works by recognizing character patterns.