Simple bash script to make searchable PDF's with Tesseract OCR (adds a hidden text layer underneath the original image).

The script takes 1 argument: the directory in which your PDF's are stored:

./ocr.sh /path/to/directory/with/pdfs

It will output searchable PDF's in the same directory, with an ocr- prefix.

This script requires tesseract and ghostscript, which on a Mac can be installed with Homebrew:

brew install brew install gs tesseract tesseract-lang

• The output PDF files are much bigger than the input ones
• No multithreading, 1 file at a time