# OCR Service

A modular Python service for performing OCR on various documents (Aadhaar, PAN, E-Bills, etc.).

## Project Structure

```text
ocr_service/
├── app_ocr.py              # FastAPI entry point
├── src/
│   ├── common/             # Shared utilities (logging, etc.)
│   ├── core/               # Document-specific OCR logic
│   │   └── aadhaar.py      # Aadhaar card processing
│   └── utils/              # Helper modules
│       ├── address_parser.py
│       ├── image.py        # Image cleaning & transformation
│       └── ocr_engines.py  # Pytesseract & EasyOCR wrappers
├── ocr_uploads/            # Temporary storage for uploads
└── requirements.txt
```

## How to Extend

To add a new document type (e.g., `pan_ocr`):

1. **Create a Core Module**: Add `src/core/pan.py`.
2. **Implement Logic**: Use the utilities in `src/utils/` for image preprocessing and OCR extraction.
3. **Add Endpoint**: Update `app_ocr.py` to include a new endpoint or update the existing one to route to your new core module.

## Setup

1. Install dependencies:
   ```bash
   pip install -r requirements.txt
   ```
2. Create a `.env` file with necessary API keys:
   - `OPENAI_API_KEY`
   - `GEMINI_API_KEY`
   - AWS credentials for CloudWatch logging.

3. Run the service:
   ```bash
   python app_ocr.py
   ```