Over-Speed and License Plate Detection of Vehicles

Abstract

With the swift expansion of the global economy, cities in various nations may face day to day problems like road congestion, frequent accidents, deterioration of the traffic conditions, or other urban traffic concerns. Vehicle detection technology based on video can collect a wealth of information from video frame sequences, such as vehicle speed, vehicle type, and vehicle number plate, at a cheap cost and with great efficiency. These electronic technologies are not only useful in people’s daily lives, but they also provide management with safe and efficient services. If we solely rely on human resources, such as law enforcement officers, we may face numerous issues, including high costs and low efficiency. We have built an integrated system for speed and license plate detection of vehicles. In the method presented here, the vehicle is first segmented and extracted from a video feed, using YOLOv5 algorithm. Next, the speed of the car is calculated using a simple algorithm and then the license plate snapshots are detected. Finally, optical character recognition is applied on the license plate image. This paper presents a thorough analysis of the cutting-edge approaches for detecting and recognising vehicles, their speeds and number plates.