# Core dependencies
Pillow>=10.0.0
pdf2image>=1.16.0

# OCR engines (choose one)
# pytesseract>=0.3.10  # Local OCR
# google-cloud-vision>=3.4.0  # Google Cloud Vision
# boto3>=1.28.0  # AWS Textract

# PDF processing
PyPDF2>=3.0.0

# Optional: LLM for entity extraction
# openai>=1.0.0
