Traffic Signal Violation Detection Using Machine Learning

Overview

This project implements an automated Traffic Signal Violation Detection System using machine learning techniques. The system leverages computer vision algorithms, specifically the YOLO (You Only Look Once) framework, for real-time detection of vehicles violating traffic signals.

Key Features

• Data Collection: Collected traffic data from road intersections and traffic cameras for building the dataset.
• Preprocessing: Data preprocessing steps to clean and prepare the dataset for model training.
• Machine Learning Model: Developed a YOLO-based model for real-time detection of vehicles that violate traffic signals.
• Violation Detection: Integrated license plate recognition using Tesseract OCR to identify vehicles and automatically issue fines.
• Web Deployment: Created a web interface for monitoring violations and viewing detection results.
• Dashboard: Provided a dashboard for visualizing traffic violation trends and system performance.

Technologies Used

• Machine Learning: Python, TensorFlow, YOLOv5
• Object Detection: YOLO for real-time vehicle and signal detection
• Web Development: HTML, CSS, JavaScript, Flask

How to Run

1. Clone the repository:

   git clone https://github.com/aryanjadav037/traffic-signal-violation-detection.git
   cd traffic-signal-violation-detection

2. Run the Flask app:

   python main.py

Contributions

Feel free to open issues or pull requests if you find any bugs or have suggestions for improvements.

License

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