A tool to OCR a PDF (or supported images) and add a text "layer" (a "pdf sandwich") in the original file making it a searchable PDF. The script uses only open source tools.
This software is free, but if you like it, please donate to support new features.
Bitcoin (BTC) address: 173D1zQQyzvCCCek9b1SpDvh7JikBEdtRJ
Tips are also welcome!
Dogecoin (DOGE) address: D94hD2qPnkxmZk8qa1b6F1d7NfUrPkmcrG
PIX (Brazilian Instant Payments): 0726e8f2-7e59-488a-8abb-bda8f0d7d9ce
Please contact for donations and tips in other cryptocurrencies.
In Linux, installation is straightforward. Just install required packages and be happy. You can use "install_command" script to copy required files to "/usr/local/bin".
In macOS, you will need macports.
# First install Xcode from Mac App Store, then:
xcode-select --install
sudo xcodebuild -license
# Install Macports from https://www.macports.org/install.php
sudo port selfupdate
# Install tesseract as main ocr engine (Portuguese included below - please add your preferred languages)
sudo port install git libtool automake autoconf tesseract tesseract-por tesseract-osd tesseract-eng
# Install cuneiform (the optional ocr engine - see flag "-c")
sudo port install cuneiform
# Install qpdf (optional for better performance)
sudo port install qpdf
# Install python 3 and other dependencies
sudo port install python39 py39-pip poppler poppler-data ImageMagick ghostscript
# Configure default python3 installer
sudo port select --set python python39
sudo port select --set python3 python39
sudo port select --set pip pip39
sudo port select --set pip3 pip39
# Configure venv and python deps in fixed home directory
python3 -m venv ~/pdf2pdfocr-venv
~/pdf2pdfocr-venv/bin/python3 -m pip install --upgrade pip
~/pdf2pdfocr-venv/bin/pip3 install --upgrade setuptools
~/pdf2pdfocr-venv/bin/pip3 install -r requirements.txt
~/pdf2pdfocr-venv/bin/pip3 install -r requirements_gui.txt
# Copy main scripts to venv
cp pdf2pdfocr.py pdf2pdfocr_gui.py pdf2pdfocr_multibackground.py ~/pdf2pdfocr-venv/bin
sudo ./install_command
Cuneiform and qpdf are optional.
In Windows, you will need to manually install required software. Please read "install_windows.txt" file and try the tutorial with scoop tool. It's easy! :-)
The Dockerfile can be used to build a docker image to run pdf2pdfocr inside a container. To build the image, please download all sources and run.
docker build -t leofcardoso/pdf2pdfocr:latest .
It's also possible to pull the docker image from docker hub.
docker pull leofcardoso/pdf2pdfocr
You can run the application with docker run.
docker run --rm -v "$(pwd):/home/docker" leofcardoso/pdf2pdfocr -v -i ./sample_file.pdf
This will create a searchable (OCR) PDF file in the same dir of "input_file".
pdf2pdfocr.py -i <input_file>
In some cases, you will want to deal with option flags. Please use:
pdf2pdfocr.py --help
to view all the options.
It's also possible to use GUI.
pdf2pdfocr_gui.py <<optional input file>>
Caseiro com orgulho! ;-)