License Plate Recognition and Tracking

This project utilizes computer vision techniques to perform license plate recognition and vehicle tracking in a video. It uses the YOLOv8n model for vehicle detection and a custom model for license plate detection. The project is designed to process a video file named "sample.mp4" located in the selected folder.

Setup

1. Clone the repository.
2. Install the required packages listed in requirements.txt.
3. Clone the 'sort' repository into the execution folder of the scripts from the following link: https://github.com/abewley/sort.git.
4. Ensure the selected folder contains a video file named "sample.mp4".

Usage

1. Run the main.py script.
2. Select the folder containing the video file.
3. The program will process the video, detecting vehicles, tracking them, and recognizing license plates.
4. The results will be saved in a CSV file named test_interpolated.csv and an output video file named out.mp4.

Results

The program outputs an annotated video with bounding boxes around vehicles and their respective license plates. The license plate numbers are also displayed on the video.

Requirements

• Python 3.7+
• OpenCV
• NumPy
• pandas
• sort (from util)

Notes

• The program expects a specific video file name ("sample.mp4") in the selected folder. Please ensure the file is present before running the program.
• The util module contains helper functions for processing frames, drawing borders, and managing license plate recognition.