rrrlw / read-pdf

Convert PDFs of scanned images of text into TXT format so that text functions (e.g. search) can be used.

read-pdf: Convert PDF of scanned text into TXT format

Raoul R. Wadhwa, Marvin R. Natowicz

Cleveland Clinic Lerner College of Medicine

Case Western Reserve University

---

Purpose

Working with large PDF documents composed of scanned pages of text documents can be daunting, even for simple tasks such as searching for a word (Ctrl+F doesn't work!). read-pdf uses qpdf, ImageMagick, and Tesseract OCR to do all the heavy lifting (in fact, read-pdf is just a short wrapper of these tools).

Requirements

• qpdf
• ImageMagick
• Tesseract OCR

Use

1. Create a directory that satisfies the following requirements:
   • contains read-pdf.sh
   • contains the PDF file you would like to convert to TXT format
   • does not contain a directory named temp (used to extract and process individual pages)
2. Using Terminal, cd into this directory and run the following, where Input.pdf should be replaced by name of the PDF you would like to convert, 10 should be replaced by the number of pages in Input.pdf, and Output.txt should be replaced by the desired name of the file that will contain the converted PDF in TXT format:

./read-pdf.sh Input.pdf 10 Output.txt

Example

The following code can be directly run on Terminal to convert the first 5 pages of the TestRecord.pdf sample file included on read-pdf's [GitHub repo] to TXT format. Note: the code will not run correctly unless you have all of the listed requirements installed.

git clone https://github.com/rrrlw/read-pdf.git
cd read-pdf
./read-pdf.sh TestRecord.pdf 5 output.txt

Contribute

To contribute to this repository, please fork it and submit a pull request with modifications. For ease of review, please also include what you changed and why.