Automatic Vehicle Classification for Electronic Toll Collection using YOLOv3

Abstract

This paper proposes the application of object detection and classification application in the area of intelligent transportation system, that is in Electronic Toll Collection (ETC) system. ETC is a cashless toll collection system, where there should be no human being involved at the toll booth anymore. Due to different toll rate applies to different type of vehicle, there is a need to have a system to detect and classify the type of vehicle passing through the toll booth. Thus, this paper proposed an automatic vehicle classification using an algorithm based on YOLOv3. The proposed model is tested and validated by extensive experiments. The result shows that the propose model manage to identify the type of vehicle in more than 92.12% accurate.