Helmet and Number Plate Detection using Deep Learning

PROJECT ABSTRACT:

This project tackles road safety and enforcement challenges using cutting-edge deep learning. The system, built with Python, leverages the swift and precise YOLOv8 object detection model for identifying helmets and license plates in real-time. A user-friendly interface, constructed with HTML, CSS, and JavaScript, is powered by the Flask framework for seamless interaction.

To ensure adaptability to diverse environments, the YOLOv8 model is trained on a rich dataset encompassing helmets and license plates in various orientations. The model's impressive training (88.00%) and validation (79.00%) accuracy signifies its ability to handle real-world complexities.

The system offers three distinct modes:

Image Analysis: Upload and analyze static images to detect helmets and license plates.
Video Processing: Examine pre-recorded videos, extracting frames and applying YOLOv8 to identify objects throughout the sequence.
Live Camera Feed: Utilize a webcam for real-time detection, ideal for dynamic monitoring.

This versatility caters to various applications, from static analysis to live traffic surveillance. The Flask-powered backend ensures smooth data handling, while the front-end prioritizes user experience.

In essence, this project showcases the power of deep learning in promoting road safety and regulatory adherence. By automating helmet detection and vehicle identification, it offers a practical tool for traffic authorities to effectively monitor and enforce safety regulations. The model's high accuracy during training and validation paves the way for successful real-world implementation of YOLOv8.

PROJECT OUTPUT VIDEO:

REFERENCE:

PEIJIAN JIN, HANG LI , WEILONG YAN, AND JINRONG XU, “YOLO-ESCA: A High-Performance Safety Helmet Standard Wearing Behavior Detection Model Based on Improved YOLOv5”, IEEE Access ( Volume: 12), 2024.

Helmet and Number Plate Detection using Deep Learning | Python Project