Getting started

The following Jupyter notebooks should walk you through all the key capabilities of Dawsonia.

• Check your installation
• Table detection with SCIPY_PROJ method
  • Input data and table formats configuration
  • Demo: Table detection
• Table detection with OPENCV_CONTOURS
  • Demo: Table detection
• Digitization
  • Demo: digitization
  • Results
    • Preprocessed Image
    • Digitized Text
    • Probabilities
    • Statistics
    • Table Meta
• HTR model inference
  • Inference
    • CVL-strings dataset
    • From SMHI’s training data
  • Conclusion

Once you have acquainted with Dawsonia, take a look at the user guide and the API documentation for a deep dive.