yapot

Yet Another PDF OCR Tool

This is a library (tool) that makes PDF -> Text as easy as possble by doing a lot of the hard stuff for you!

You will need ImageMagick, Tesseract, and QPDF to use yapot.

Ubuntu
------
> sudo apt-get install imagemagick libmagickcore-dev
> sudo apt-get install tesseract-ocr
> sudo apt-get install qpdf

To use yapot, do the following:

> pip install yapot

Then some code:

from yapot import convert_document

success, pdf_text = convert_document('file.pdf')

if success == True:
    with open('file.txt', 'w') as f:
        f.write(pdf_text)
else:
    print "Unable to convert PDF!"

It's that simple!

Some more advanced things you can do are set the resolution, page delineation, and tell yapot not to delete temporary files (this can be useful when debugging nasty pdf's).

success, pdf_text = yapot.convert_document(
    pdf_filename = pdf_filename,       # The name of the pdf file
    resolution = 200,                  # Image DPI resolution
    delete_files = True,               # delete temporary files
    page_delineation = '\n--------\n', # page deination text
    verbose = False,                   # output verbosity
    temp_dir = str(uuid.uuid4()),      # location of temp directory to use
    password = '',                     # password for PDF file
    make_thumbs = True,                # create thrumbnails for each page
    thumb_size = 512,                  # width of thumbnail image
    thumb_dir = './thumbs',            # directory to place thumbnails
    thumb_prefix = 'thumb_page_',      # prefix for thumbnail images
)