Basic Python Script for running Tesseract OCR on PDFs

convert.py

import os			    # for magick and tesseract commands
import time			  # for epoch time
import calendar 	# for epoch time
from PyPDF2 import PdfFileMerger

dir_files = [f for f in os.listdir(".") if os.path.isfile(os.path.join(".", f))]
epoch_time = int(calendar.timegm(time.gmtime()))
print(dir_files)

for file in dir_files: # look at every file in the current directory
	if file.endswith('.pdf'): # if it is a PDF, use it
		print('Working on converting: ' + file)
		# setup
		file = file.replace('.pdf', '') # get just the filepath without the extension
		folder = str(int(epoch_time)) + '_' + file # generate a folder name for temporary images
		combined = folder + '/' + file # come up with temporary export path
		# create folder
		if not os.path.exists(folder): # make the temporary folder
			os.makedirs(folder)
		# convert PDF to PNG(s)
		magick = 'convert -density 150 "' + file + '.pdf" "' + combined + '-%04d.png"'
		print(magick)
		os.system(magick)
		# convert PNG(s) to PDF(s) with OCR data
		pngs = [f for f in os.listdir(folder) if os.path.isfile(os.path.join(folder, f))]
		for pic in pngs:
			if pic.endswith('.png'):
				combined_pic = folder + '/' + pic
				print(combined_pic)
				tesseract = 'tesseract "' + combined_pic + '" "' + combined_pic + '-ocr" PDF'
				print(tesseract)
				os.system(tesseract)
		# combine OCR'd PDFs into one
		ocr_pdfs = [f for f in os.listdir(folder) if os.path.isfile(os.path.join(folder, f))]

		merger = PdfFileMerger()
		for pdf in ocr_pdfs:
			if pdf.endswith('.pdf'):
				merger.append(folder + '/' + pdf)

		merger.write(file + '-ocr-combined.pdf')
		merger.close()

@BromTeque
hi , I want to run the script for pdf pages containing both english and a non-english language. i realized it can be done using the '+' parameter between the language codes, however it also assigns relative priority to the languages (based on which is mentioned before & after the + symbol).. the way to resolve that is probably using langdetect but I can't figure out how to code that .. can you please help me ?

@coder-curious

I don't think I'll be able to help you. I'm a very busy individual at the moment. It sounds like you almost got it. A bit more tinkering and you'll probably figure it out. Good luck!

Thank you. The video is very helpful

What if I want txt instead of PDF?

Thanks again.

Hi, how do i add . tif files to your code with Imagemagick?

Hey guys, pay attention to the command pdffilemerger,it is deprecated.
Change "from PyPDF2 import PdfMerger" to "from PyPDF2 import PdfMerger" and "PdfFileMerger()" to "PdfMerger()"

getting this error
PS D:> python ./convert.py
['1.pdf', 'convert.py']
Working on converting: 1.pdf
convert -density 150 "1.pdf" "1738563963_1/1-%04d.png"
Invalid Parameter - 150

getting this error PS D:> python ./convert.py ['1.pdf', 'convert.py'] Working on converting: 1.pdf convert -density 150 "1.pdf" "1738563963_1/1-%04d.png" Invalid Parameter - 150

apparently the new way to use imagemagick is
magick convert -density 150 "1.pdf" "1738563963_1/1-%04d.png"
instead of
convert -density 150 "1.pdf" "1738563963_1/1-%04d.png"

in line 18 of the script should be
magick = 'magick convert -density 150 "' + file + '.pdf" "' + combined + '-%04d.png"'

if it still shows error try just removing the convert and use only 'magick -density....etc'