Use Tesseract to add a text layer to a PDF via OCR

(Using macOS)

Dependencies:

• ImageMagick
• Tesseract
• pdfmerge

curl -OL https://github.com/gsauthof/utility/raw/master/pdfmerge.py
pip install pdfrw
brew install imagemagick
brew install tesseract

Assuming the source file is called report.pdf:

Convert the PDF into one PNG per page:

convert -density 150 report.pdf +adjoin report-%03d.png

Perform OCR on each page and produce a text-only PDF called textonly.pdf:

ls report-*.png | tesseract -c textonly_pdf=1 --dpi 150 - textonly pdf

Merge the original PDF with the text-only PDF into a PDF called searchable.pdf:

python pdfmerge.py --pdfrw textonly.pdf report.pdf searchable.pdf