Digital Text Conversion in C++ Using OpenCV & Tesseract

• Overview
• Dependencies
• Additional Info
• Citations

Overview

A Cinder application that allows you to write text and convert it to standard, digital text. It also provides the user with the option to send their converted output to a file of their choice.

Dependencies

• g++/gcc 4.8 and above, clang 3.4 and above
• cmake
• OpenCV 4.5.0
• Tesseract 4.1.1

Ubuntu

$ sudo apt-get install gcc cmake opencv tesseract

• Can be built using CLion

MacOS

$ brew install gcc cmake opencv tesseract

• Can be built using CLion

Windows & Other Environments

• Check out the links above for more info on how to install dependencies for your system. You will need MSVC-2015 or higher to build this project on Windows.

Installation

• After obtaining the necessary dependencies, clone the repository with git clone https://github.com/shruthikmusukula/DigitalTextConverter.git on your local machine.

Code Style

This codebase was written in accordance with the Google C++ Style Guide (https://google.github.io/styleguide/cppguide.html). Code from the OpenCV and Tesseract frameworks may not adhere to these guidelines.

Demo

Below is a full demo of the application and its major features.

1) Live Character Recognition

2) Colorful User Input

3) Write To A File

4) Boost CLI Integration for Image to Text Conversion

By passing in appropriate image input paths and textfile output paths, a user can translate the text in an image of their choice to digital text.

Additional Info

Here are some additional resources if you are looking to build upon this codebase or encounter any errors:

A Common Error

Failed loading language 'eng'
Tesseract couldn't load any languages!
Could not initialize tesseract.

Please refer to this Github Issue. You need to add this training data file to your tesseract installation. On Ubuntu, download and place this file at /usr/share/tesseract-ocr/4.1.1/tessdata/eng.traineddata. On MacOS, download and place this file at /usr/local/Cellar/tesseract/4.1.1/tessdata/eng.traineddata.

Improving the Quality of the Tesseract Engine

Excerpt from Tesseract Docs:

Tesseract does various image processing operations internally (using the Leptonica library) before doing the actual OCR. It generally does a very good job of this, but there will inevitably be cases where it isn’t good enough, which can result in a significant reduction in accuracy.

Please look here for additional information.

Citations

All necessary citations are listed wherever applied in the codebase.