Helmet Detection and Number Plate Recognition Using Machine Learning

Abstract

The continuous mobilization of vehicles has led to a surge in the number of road accidents across the world. To get better at this, the government is trying to focus on the safest and most preventive measures in traffic. The practice of direct observation is found to be time taking and a lot of human effort is needed. So, our main idea is to introduce a helmet and license plate detection mechanism. This project attempts to implement a detection process through a few machine-learning algorithms by using predefined libraries. This system notices a person with/without a helmet thereby imposing fines on the detected candidate’s license plate. Further, this research work concludes that the automatic identification of helmets can overcome the challenges faced by the manual data collection process. Moreover, this project work assumed that, through data collection, the algorithm can help to track helmet use and promote its active use by people to ensure road safety.