Paving the Way to Safety: An Automated Traffic Management and Challan Generation System

In India, traffic regulations have become a major concern due to various problems that have arisen. One of the most significant issues is the rising number of accidents and deaths caused by people riding motorcycles and mopeds without helmets. The existing system relies on Closed Circuit Television (CCTV) recordings, where traffic police must manually monitor violations and zoom in to identify the license plate number of the rider. However, this approach is time consuming and requires a lot of manpower, especially since the number of motorcycle riders increasing every day. To address this issue, researchers have developed a new system that can automatically detect violations of not wearing a helmet while riding a motorcycle pr moped. This system utilizes advanced features like Convolutional Neural Networks, Local Binary Patterns (LBP), Region based Convolutional Neural Networks (R-CNN), Histogram Oriented Gradients (HOG) and HAAR to identify and extract the license number plate of the offending vehicle. Given that motorcycle accidents are a leading cause of fatalities on the roads, it is crucial to enforce the use of helmets. The government has made it a punishable offense and law enforcement agencies have been using manual methods to catch violators. However, the proposed approach uses video surveillance to detect violators and record their license plate numbers, thereby making the process more efficient and more effective. The system will generate an E-challan containing the details of the offender, and also create a database with records of all offenders to ensure accurate identification. To identify different types of helmets with minimal computational cost, the system relies solely on pure machine learning techniques.