RotDams/JustReadIT

JustReadIt

JustReadIt is a text recognition tool developed as part of EPITA's 3rd semester project.

This software allows to extract a text from an image, and this through 3 processes.

• A segmentation of the image into paragraphs, lines, words and characters.

• A recognition of each character through a neural network.

Getting Started

These instructions will get you a copy of the project up and running on your machine for testing purposes and production.

Prerequisites

• SDL 1

• GTK 3

• SDL_image

• SDL_gfx

Installing

• Download the project from github

$ git clone https://github.com/RotDams/JustReadIT

• Go in the created folder, make and run

$ cd JustReadIT
$ make

• A JustReadIt executable should be created.

Running

Our program currently offers two main functions, segmentation and xor.

Complete OCR

The complete OCR can be start easily, and help you by showing an UI. You can start it by just typing:

$ ./JustReadIt

Trainning

If you want, you can try the neural network. For it, just start the program with the -t flag.

$ ./JustReadIt -t

Our function can also take 3 optional parameters.

• --save (or -s): if specify, the code will save the data of neural networks after working.
• --load (or -l): if specify, the code will load the saved data of neural networks before working.
• --fonts x (or -f x): if specify, the code will set the number of font to x.

Credits

JustReadIt was developed by damien.champeyroux, quentin.briolant, adrien.barens and sami.beyhum.