python-ocr-pdftotext

Converts scanned PDF files to plain text using Tesseract OCR.

How it works

1. PDF → individual JPEG frames (via pdf2image)
2. Each frame → Tesseract OCR → raw text
3. Text chunks assembled into a single output file per PDF

Project layout

.
├── input/          # place input PDF files here
├── temp/           # intermediate JPEGs (auto-cleaned after run)
├── output/         # extracted .txt files land here
├── ocr.py          # main script
└── README.txt      # setup notes and common errors

Requirements

sudo apt-get install tesseract-ocr poppler-utils   # Linux
brew install tesseract poppler                      # macOS
pip install pytesseract pdf2image Pillow

Usage

# Place PDFs in input/, then:
python3 ocr.py
# Output .txt files appear in output/

Notes

• Works best on scanned documents with clean, high-contrast text
• Multi-column layouts may need post-processing
• Language pack: defaults to English (eng); set TESSDATA_PREFIX for others