django-ocr 0.0.0

A lightweight Django-based OCR package

django-ocr

A lightweight Django-based OCR package that extracts text from images and allows dynamic mapping of extracted text into structured dictionaries.

Features

• Extracts text from images using Tesseract OCR
• Supports dynamic key-value mapping for structured text output
• Simple API for easy integration into Django projects
• Fully customizable mapping to extract expected information

Installation

You can install django-ocr using pip:

pip install django-ocr

Dependencies

Make sure you have Tesseract-OCR installed on your system. If not, install it using:

• Ubuntu/Debian:

  sudo apt install tesseract-ocr
  

• MacOS:

  brew install tesseract
  

• Windows: Download and install from Tesseract's official site.

Usage

1. Extract Text from an Image

from django_ocr import process_image

text = process_image("sample_image.png")
print(text)  # Output: Extracted text from the image

2. Extract and Map Text to Structured Dictionary

If you have an image containing structured information, such as:

Name: John
Last Name: Doe

You can map it dynamically:

from django_ocr import process_image_with_mapping

image_path = "sample_image.png"
mappings = {
    "first_name": ["Name"],
    "last_name": ["Last Name"]
}

result = process_image_with_mapping(image_path, mappings)
print(result)
# Output: {'first_name': 'John', 'last_name': 'Doe'}

API Reference

process_image(image_path: str) -> str

Extracts text from an image and returns it as a string.

process_image_with_mapping(image_path: str, mappings: dict) -> dict

Extracts text from an image and maps values to a structured dictionary based on provided mappings.

• image_path: Path to the image file
• mappings: A dictionary where keys represent desired output fields, and values are lists of keywords to search for

Customization

You can define your own mappings dynamically based on the expected text format.

Example for extracting an ID and an email:

mappings = {
    "user_id": ["ID", "User ID"],
    "email": ["Email"]
}

Contributing

Feel free to open issues or submit pull requests to improve this package!

License

MIT License