Practical integration of RFID and image-based vehicle identification data to detect illegal vehicles

Abstract

Number plate cloning has become increasingly prevalent due to the ease with which criminals can duplicate plates to impersonate legitimate vehicles and evade road camera systems. A previous paper introduced a method that combines RFID with computer vision to reduce such illegal road use. This is achieved by tagging each vehicle number plate with an RFID chip that stores the different features of the vehicle, allowing data extracted by the RFID reader to be compared with information derived from the camera feed. To prevent cloning, a digital signature using public and private keys ensures that the tag data originates from an authorized source. The authorities can leverage this information to remove such vehicles from the road without affecting the legal road user’s experience as it enables immediate intervention on a tactical level without disrupting the flow of legal vehicles. This paper describes the practical evaluation of this system in various test scenarios. The integrated solution synchronizes the RFID and camera data streams within Docker, ensuring parallel processing without data loss as vehicles pass through the system. Testing demonstrated high tag read reliability across speeds from 10 km/h to 70 km/h, with data accurately matched under various conditions. These results show that this combined approach effectively strengthens the verification of vehicle identities, making plate cloning more challenging by introducing a secure, authentication system.