MACHINE LEARNING BASED VISION SYSTEM FOR AUTOMATED NUMBER PLATE MONITORING AND VIOLATION DETECTION

Computer Science and Engineering

ABSTRACT Utilizing the powerful OpenCV computer vision library, this project presents a holistic solution aimed at advancing road safety through real-time helmet detection, license plate identification, and a unique feature for recognizing stolen vehicles. The primary objective is to enhance road safety by detecting helmet-wearing motorcyclists and identifying car license plates in real-time, contributing to a safer driving environment. Notably, the project incorporates a data input component that efficiently stores detected helmet statuses, license plate information, and warns against stolen vehicles in an Excel spreadsheet, facilitating easy tracking and analysis. A significant highlight of this project is its efficient and precise helmet identification capacity, achieved through the Python-based YOLOv3 (You Only Look Once version 3) deep learning framework. This method is pivotal for recognizing individuals wearing helmets across various settings, such as construction sites or sporting events, where helmet use is mandatory. YOLOv3's real-time object identification capabilities make it well-suited for this application, demonstrating consistent performance across diverse helmet types, angles, and lighting conditions. In addition to helmet detection, the project introduces an innovative approach to license plate recognition (LPR) by seamlessly connecting with third-party APIs. The primary objective is to detect license plates in real-time using external LPR services, enhancing plate identification and character recognition accuracy. This approach broadens the system's applications to areas like parking management, security, and traffic monitoring. To further augment the system's capabilities, a new feature has been incorporated to detect stolen vehicles. When a stolen vehicle's license plate is identified, the system provides a warning, reinforcing its role in both data collection and traffic safety. This feature aligns with law enforcement efforts and underscores the project's contribution to technological advancements in road safety. Keywords: YOLOv3, real-time, OpenCV, APIs, License Plate Recognition

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