# OCR API

A FastAPI-based OCR (Optical Character Recognition) service that can extract text from PDF files and images using Tesseract.

## Features

- **PDF Processing**: Convert PDF files to images and extract text using OCR
- **Image Processing**: Direct OCR processing of image files (PNG, JPG, JPEG, BMP, TIFF)
- **Multi-language Support**: Supports English, French, German, Spanish, and Italian
- **Confidence Scoring**: Returns confidence scores for OCR results
- **RESTful API**: Clean, documented API endpoints
- **Docker Support**: Easy deployment with Docker and Docker Compose
- **File Validation**: Built-in file type and size validation
- **Background Processing**: Efficient file cleanup and processing

## Quick Start

### Using Docker (Recommended)

1. **Clone and build the container:**

   ```bash
   docker-compose up --build
   ```

2. **Access the API:**
   - API Documentation: http://localhost:8000/docs
   - Health Check: http://localhost:8000/health

### Local Development

1. **Install system dependencies:**

   ```bash
   # Ubuntu/Debian
   sudo apt-get update
   sudo apt-get install tesseract-ocr tesseract-ocr-eng poppler-utils libpoppler-cpp-dev pkg-config

   # macOS
   brew install tesseract poppler

   # Arch Linux
   sudo pacman -S tesseract tesseract-data-eng poppler
   ```

2. **Install Python dependencies:**

   ```bash
   pip install -r requirements.txt
   ```

3. **Run the application:**
   ```bash
   python main.py
   ```

## API Endpoints

### Health Check

- **GET** `/health` - Check if the service is running

### PDF OCR

- **POST** `/ocr/pdf` - Extract text from PDF files
  - **File**: PDF file (max 50MB)
  - **Response**: Extracted text with metadata

### Image OCR

- **POST** `/ocr/image` - Extract text from image files
  - **File**: Image file (PNG, JPG, JPEG, BMP, TIFF)
  - **Response**: Extracted text with metadata

## Usage Examples

### Using curl

**Process a PDF file:**

```bash
curl -X POST "http://localhost:8000/ocr/pdf" \
     -H "accept: application/json" \
     -H "Content-Type: multipart/form-data" \
     -F "file=@document.pdf"
```

**Process an image file:**

```bash
curl -X POST "http://localhost:8000/ocr/image" \
     -H "accept: application/json" \
     -H "Content-Type: multipart/form-data" \
     -F "file=@image.png"
```

### Using Python

```python
import requests

# Process PDF
with open('document.pdf', 'rb') as f:
    files = {'file': f}
    response = requests.post('http://localhost:8000/ocr/pdf', files=files)
    result = response.json()
    print(f"Extracted text: {result['text']}")
    print(f"Processing time: {result['processing_time']}s")
    print(f"Confidence: {result['confidence']}%")
```

### Using JavaScript/Fetch

```javascript
// Process PDF
const formData = new FormData();
formData.append("file", fileInput.files[0]);

fetch("http://localhost:8000/ocr/pdf", {
  method: "POST",
  body: formData,
})
  .then((response) => response.json())
  .then((data) => {
    console.log("Extracted text:", data.text);
    console.log("Processing time:", data.processing_time);
    console.log("Confidence:", data.confidence);
  });
```

## Response Format

```json
{
  "text": "Extracted text content...",
  "pages": 3,
  "processing_time": 2.45,
  "confidence": 85.2
}
```

## Configuration

### Environment Variables

- `PYTHONUNBUFFERED=1` - Ensures Python output is not buffered
- `PYTHONDONTWRITEBYTECODE=1` - Prevents Python from writing .pyc files

### Tesseract Configuration

The Docker image includes multiple language packs:

- English (eng)
- French (fra)
- German (deu)
- Spanish (spa)
- Italian (ita)

To add more languages, modify the Dockerfile and add additional `tesseract-ocr-*` packages.

## Performance Tips

1. **Image Quality**: Higher DPI (300) provides better OCR accuracy but slower processing
2. **File Size**: Larger files take longer to process
3. **Text Quality**: Clear, high-contrast text yields better results
4. **Language**: Specify the correct language for better accuracy

## Troubleshooting

### Common Issues

1. **Tesseract not found**: Ensure Tesseract is installed on your system
2. **PDF processing errors**: Check if Poppler utilities are installed
3. **Memory issues**: Reduce DPI or process smaller files
4. **Poor OCR quality**: Ensure images are clear and have good contrast

### Logs

The application logs processing information. Check Docker logs:

```bash
docker-compose logs ocr-api
```

## Development

### Project Structure

```
ocr-api/
├── main.py              # FastAPI application
├── requirements.txt     # Python dependencies
├── Dockerfile          # Docker configuration
├── docker-compose.yml  # Docker Compose configuration
└── README.md          # This file
```

### Adding Features

1. **New OCR engines**: Modify the `process_pdf_with_ocr` function
2. **Additional file formats**: Add new endpoints and processing logic
3. **Custom preprocessing**: Add image preprocessing steps before OCR
4. **Batch processing**: Implement queue-based processing for multiple files

## License

This project is open source and available under the MIT License.