Table of contents

• General info
• Requirements
• Technologies
• Features
• Installation
• Screenshots

General info

Application which aims to visualise how OCR work (using tessaract OCR library and its HOCR output format).

Requirements

• PHP 7.3+
• mySQL Database
• Tesseract Open Source OCR Engine - https://github.com/tesseract-ocr/tesseract
• ImageMagick
• PHP imagick module

Technologies

Project is created with:

• Symfony 4
• Tesseract OCR for PHP - https://github.com/thiagoalessio/tesseract-ocr-for-php
• Imagick/ImageMagick
• PHPHtmlParser - https://github.com/paquettg/php-html-parser

Features

• Displaying tessaractOCR bounding boxes on image (words, lines, paragraphs)
• Displaying recognised phrases over text on image
• Customisation of draw parameters - changing bounding box stroke color, font-size etc

Installation

Clone the repository

Install front-end dependencies

yarn install

Compile assets

yarn encore dev

Run composer

composer install

Create database

 php bin/console doctrine:database:create

Run migrations

php bin/console doctrine:migrations:migrate

Screenshots

Homepage

Image with all bounding boxes

Part of the image with recognised text