PDF Text Extraction Tool

A Python utility for extracting text from scanned PDF documents using OCR technology.

Description

This tool uses Tesseract OCR to extract text from PDF files, saving each page as a separate text file. It can also combine all extracted pages into a single document.

Key features:

• Converts PDF pages to images
• Applies image preprocessing to improve OCR accuracy
• Extracts text using Tesseract OCR
• Combines extracted text files into a single document (optional)
• Automatically sets up a virtual environment with required dependencies

Installation

This script is designed to be run directly and is not installable as a package. It automatically creates a dedicated virtual environment with all required Python dependencies (pdf2image, pytesseract, pillow) the first time it runs.

Requirements

• Python 3.6+
• Tesseract OCR must be installed on your system
  • Ubuntu/Debian: sudo apt install tesseract-ocr
  • macOS: brew install tesseract
  • Windows: Download installer from https://github.com/UB-Mannheim/tesseract/wiki

Usage

python pdf_text_extraction.py --pdf_path <path_to_pdf> --output_dir <output_directory> [--combine] [--skip-extraction]

Arguments

• --pdf_path: Path to the PDF file to process
• --output_dir: Directory to save extracted text files
• --combine: (Optional) Combine all extracted text files into a single document
• --skip-extraction: (Optional) Skip extraction and only combine existing text files

Examples

Extract text from a PDF:

python pdf_text_extraction.py --pdf_path document.pdf --output_dir ./extracted_text

Extract text and combine into a single document:

python pdf_text_extraction.py --pdf_path document.pdf --output_dir ./extracted_text --combine

Only combine existing text files:

python pdf_text_extraction.py --output_dir ./extracted_text --combine --skip-extraction

Created by

This tool was vibe coded by Erika Rice Scherpelz using Claude Code.