Automatic Number Plate Recognition (ANPR) Project

Project Structure

• Django Application: The Django folder is located within the ANPR subfolder.
• Datasets:
  • Training dataset: new_tr_dataset_ref
  • Validation dataset: new_tr_valid_dataset_ref
  • Test dataset: new_tr_test_dataset_ref
• Model Weights:
  • Custom model weights: best_tr_model2 (located within the ANPR subfolder)
  • VGG16 transfer learning model weights: best_tr_pretrained_model
• Development Notebook: The notebook used to develop the character recognition model is reworked_tr_model.ipynb (developed on Google Colab).
• Dependencies: The virtual environment for this project was deleted due to its size. The project prerequisites can be found in requirements.txt within the ANPR subfolder. This file was created using pip freeze > requirements.txt.

Running the Server

To run the server, open a terminal or command prompt, navigate to the ANPR subfolder, and execute the following command:

python manage.py runserver

Examples