What the project do?

The class is used to classify letters according to its type. Every letter is stored in a pdf file and has its code (TID) in the header of the file:

In order to classify properly the class uses 3rd party libraries to read that code and move that pdf file into a separate folder according to its type.

The logic behind the code is:

1. Uses ghostscript library to transform the pdf into a jpg file.

2. Uses imagemagick library to crop that jpg file into a smaller file containing only the TID code

3. Uses tessaract-ocr library to read that line and extract a character which represent the type.

   

   (There are 4 known types of letters: B, E, D, F)

4. Finally, you will have a folder for each type of letter, and inside this folder you will have the pdf files.

Installation instructions [Linux]:

Class was made in plain PHP but some external libraries are required to work properly.

1. Install External Libraries:

Open your terminal and use apt-get to install the following packages:

1.1. # apt-get install tesseract-ocr

1.2. # apt-get install ghostscript

1.3. # apt-get install imagemagick

2. Configure your project

You MUST set the pathToFiles parameter stored in you config.json file.

"pathToFiles": "/home/[PROYECT_FOLDER]/pdfs"

3. Place the pdf files:

Place the pdfs that you want to classify inside the folder configured in the previous step

(A couple of pdfs were provided for testing purpose)

4. Run the project:

Inside the project folder type the following command to run the class.

# php classifier-cli.php

If everything goes well you should have the pdf files inside a folder for each type of letters.