VanGuardian: School Vehicle Security and Management

VanGuardian is an innovative software solution designed to elevate the security and streamline the management of school vehicles, offering a robust system that integrates computer vision and neural networks. The project focuses on real-time detection, identification, and verification of school vans, license plates, and specific decals known as prefixes.

Features:

1. Real-time Detection Using YOLO

• Leveraging YOLO (You Only Look Once) architecture, VanGuardian ensures efficient and accurate detection of school vans from live video feeds.

2. License Plate and Decal Recognition

• Utilizes neural networks to identify license plates and prefixes on detected vans, enhancing the system's ability to extract critical information.

3. OCR (Optical Character Recognition)

• Employs OCR technology to extract alphanumeric characters from license plates and prefixes, facilitating seamless verification.

4. List Verification

• Cross-references the extracted license plates and prefixes with a predefined list, providing instant feedback on the approval status of the school vehicle.

Getting Started:

Prerequisites:

• Ensure the installation of the required dependencies: cv2, ultralytics, PIL, pytesseract, pandas, numpy, and Tesseract OCR.

Usage:

1. Load pre-trained YOLO models for vans, license plates, and prefixes.
2. Execute the main script by running the following command in your terminal or command prompt:

python main.py

Project Background:

This project serves as the capstone for the completion of the Bachelor's degree in Computer Science. It showcases the practical application of computer vision and neural networks in addressing real-world challenges, specifically in the domain of school vehicle security and management.

Performance Metrics:

VanGuardian strives for optimal performance, with a focus on real-time processing. The average processing time for vehicle detection and verification is consistently monitored and displayed during runtime.

Contributing:

We welcome contributions and feedback! Feel free to fork the repository, open issues, and submit pull requests to enhance the functionality of VanGuardian.

License:

This project is licensed under the MIT License - see the LICENSE file for details.

Acknowledgments:

We appreciate the open-source community and the contributions of various libraries and frameworks that make VanGuardian possible.