Skip to content

Instantly share code, notes, and snippets.

@neurojojo

neurojojo/pdfimages_to_text.py

Last active March 14, 2022 18:42
Show Gist options
  • Save neurojojo/2adfc000cc72beab7e2533b382de5d47 to your computer and use it in GitHub Desktop.
Save neurojojo/2adfc000cc72beab7e2533b382de5d47 to your computer and use it in GitHub Desktop.
Download ZIP
Raw
pdfimages_to_text.py
# Variables:
# (1) pdf_as_text
# where some list elements are empty because an attempt was made to
# read in a PDF that is a scanned image (not typed text)
# (2) files
# a list of file names and paths to the PDFs
!apt-get install poppler-utils
!pip install pdf2image
from pdf2image import convert_from_path, convert_from_bytes
from IPython.display import display, Image
!pip install ocrmypdf
!pip install opencv-python
!pip3 install PIL
!pip3 install pytesseract
!pip3 install pdf2image
!sudo apt-get install tesseract-ocr
import pytesseract
def convertImagePDF( inputfile ):
inputimages = convert_from_bytes(open( inputfile , 'rb').read())
text = [ str(((pytesseract.image_to_string(inputimage)))) for inputimage in inputimages ]
return text
problem_pdfs = [ files[index] for index,this_pdf in enumerate(pdf_as_text) if len( this_pdf )==0 ]
image_pdf_text = [ convertImagePDF( problem_pdf ) for problem_pdf in problem_pdfs ]
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment