REAL TIME IMAGE SEGMENTATION FROM LIVE VIDEO STREAMING

COMPUTER SCIENCE AND ENGINEERING

This project presents a comprehensive solution aimed at enhancing road safety through real-time helmet detection and license plate recognition, utilizing the powerful OpenCV computer vision library. The primary objective of this system is to identify motorcyclists wearing helmets and recognize license plates of vehicles in real-time, thereby contributing to a safer driving environment. Additionally, the project includes a data entry component that efficiently logs detected helmet statuses and license plate information into an Excel spreadsheet, allowing for easy tracking and analysis of the collected data.One of the key highlights of this project is its efficient and accurate helmet detection capability, which is achieved through the utilization of the YOLOv3 (You Only Look Once version 3) deep learning framework implemented in Python. This approach plays a pivotal role in enhancing safety by identifying individuals wearing helmets in various contexts such as construction sites, sports events, or any scenario where helmet usage is essential. YOLOv3 is renowned for its real-time object detection capabilities, making it a suitable choice for this application.The project takes full advantage of the YOLOv3 architecture by employing a custom-trained model fine-tuned on a diverse dataset containing various helmet images. This meticulous training process ensures robust performance across different helmet types, angles, and lighting conditions. The seamless integration of the YOLOv3 model into the Python environment makes deployment and customization straightforward, enabling users to adapt the system to their specific needs.In addition to helmet detection, this project also introduces an innovative approach to license plate recognition (LPR) by seamlessly integrating with third-party APIs. The primary goal here is to accurately and efficiently recognize license plates from images and video streams in real-time, harnessing the capabilities of external LPR services. This approach not only improves the accuracy of plate detection and character recognition but also broadens the system's applications to areas like parking management, security, and traffic monitoring.The adaptability and extensibility of this project make it a valuable tool for organizations seeking efficient and accurate license plate recognition solutions. By leveraging external services, it achieves high levels of accuracy, ensuring reliable results in various real- world scenarios.In conclusion, this project represents a significant step forward in road safety and data collection. By combining real-time helmet detection and license plate recognition, it addresses critical aspects of traffic safety and management. The YOLOv3-based helmet detection and the seamless integration of external LPR services highlight its efficiency and accuracy. With its data logging capabilities, this project is not only a technological advancement but also a valuable resource for organizations and authorities striving

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.