Abstract: In this, we will discuss how traffic problems can be reduced by implementing an automatic toll collection system. This system is based on image processing and image classification. It helps in maintaining the transparency of the toll collection system. The aim is to make a digital toll collection system that will be less time-consuming and automated monitoring and control of vehicle entry-exit on a highway using machine learning techniques. At the entrance of the toll, the gate camera captures an image of the vehicle and from the image, it creates a bounding box with probability estimates of the feature classes as output. We present a review of state-of-the-art traffic monitoring systems focusing on the major functionality of vehicle classification. The result is the use of machine learning for effectively detecting vehicles in electronic tolling systems in real-time.

Keywords: e-toll, ml for vehicle classification

| DOI: 10.17148/IARJSET.2022.96155