Car plates project. OCR model (part 2/3)

This is the project for car plate OCR recognition, which include:

1. Neural network segmentation model for car plate area with number selection (part 1/3)
2. Neural network OCR model for plate character recognition (part 2/3)
3. API service for these two models (part 3/3)
4. Additional example how to use API service in Telegram bot

Dataset

Dataset include car plate crops from 27 countries, about 2 000 000 images (include two line numbers) in .jpg. Some data are bad and invalid. For more information about data and backbone selection for CRNN please see EDA.ipynb.

To download data:

make download_dataset

Environment setup

1. Create and activate python venv

   python3 -m venv venv
   . venv/bin/activate

2. Install libraries

    make install

3. Run linters

   make lint

4. Tune config.yaml and select input size of image and backbone for CRNN, see EDA notebook

5. Train

   make train

Additional information

• Inference example in notebook
• Best experiment in ClearML
• History of experiments