Help & Documentation
Complete guide to using Receipt Extractor for text extraction from receipts.
Getting Started
1. Upload Your Receipt
Click the upload area or drag and drop your receipt image. Supported formats include JPG, PNG, WEBP, and PDF files up to 100MB.
2. Select Extraction Method
Choose from 7 OCR models:
- Tesseract OCR: Fast and reliable for clear receipts
- EasyOCR: Supports 80+ languages
- PaddleOCR: High accuracy multilingual OCR
- Donut: AI-powered structured extraction
- Florence-2: Vision-language model
- CRAFT: Text detection specialist
- Spatial Multi-Method: Ensemble for maximum accuracy
3. Configure Detection Settings
Choose detection mode:
- Auto Detect: Automatic text region detection
- Manual Select: Draw custom regions
- Line Detection: Process line-by-line
- Column Mode: Handle multi-column layouts
4. Enable Preprocessing
Improve extraction accuracy:
- Deskew: Auto-rotation for straightening
- Enhancement: Image quality improvements
5. Process and Review
Click Process Receipt to extract text. Results show all detected text regions with confidence scores.
Extraction Methods Explained
| Method | Speed | Accuracy | Best Use Case |
|---|---|---|---|
| Tesseract OCR | Very Fast | 95% | Clear, high-quality receipts |
| EasyOCR | Fast | 90% | Multilingual receipts |
| PaddleOCR | Medium | 93% | Asian languages, complex layouts |
| Donut | Medium | 92% | Structured data extraction |
| Florence-2 | Slow | 94% | Challenging, low-quality images |
| CRAFT | Fast | 91% | Text region detection |
| Spatial Multi-Method | Slow | 96% | Maximum accuracy needed |
Tips for Best Results
Image Quality
- Use good lighting when photographing receipts
- Avoid shadows and glare
- Keep the receipt flat and unwrinkled
- Ensure text is in focus
Preprocessing
- Enable deskew for rotated or tilted receipts
- Use enhancement for low-quality or faded receipts
- Try different detection modes if results are poor
Model Selection
- Start with Tesseract for standard English receipts
- Use EasyOCR or PaddleOCR for non-English text
- Try Spatial Multi-Method for maximum accuracy
- Use Donut or Florence-2 for complex layouts
Troubleshooting
Poor Extraction Results
Issue: Text is missing or inaccurate
Solutions:
- Try a different extraction method
- Enable preprocessing options
- Improve image quality and re-upload
- Try manual region selection
File Upload Errors
Issue: Cannot upload receipt
Solutions:
- Check file size (max 100MB)
- Verify file format (JPG, PNG, WEBP, PDF)
- Try a different browser
- Check internet connection
Processing Timeout
Issue: Extraction takes too long
Solutions:
- Try a faster method (Tesseract)
- Reduce image size before upload
- Disable unnecessary preprocessing
- Wait and retry - server may be busy
Contact Support
Need additional help? Contact our support team:
- Email: support@receiptextractor.com
- Response Time: Within 24 hours
- Hours: Monday - Friday, 9 AM - 5 PM EST