/ocr-table

Extract tables from scanned image PDFs using Optical Character Recognition.

Primary LanguagePythonMIT LicenseMIT

ocr-table

This project aims to extract tables from scanned image PDFs using Optical Character Recognition.

Install Requirements

  1. Tesseract OCR

    sudo apt-get install tesseract-ocr

  2. Imagemagick

    sudo apt-get install imagemagick

  3. PDF Utilities

    sudo apt-get install poppler-utils

  4. Python packages

    sudo pip install -r requirements.txt

Usage

  1. Clear the pdf/ folder and copy all your pdf files to be scanned in it.

  2. Run the OCR:

    python3 shellocr.py

  3. The scanned text files shall be available in the txt/ folder once the process completes.

Alternate

  1. If the above doesn't work for you, try the alternate method.

  2. Save your file as input.pdf in the root directory.

  3. Run

    python3 pdf_miner.py