Identification and Tagging of Malicious Vehicles through License Plate Recognition

Vehicular Ad-hoc NETworks (VANET) are becoming a reality in today's world. These networks are composed of highly dynamic and capable vehicles and they rely on information that originates and is exchanged between each other. One of the main success factors of this communication is the validity of the data communicated. Hence, malicious vehicles pose a serious threat to VANETs. Once a vehicle is identified as malicious, the main challenge is to keep a centralized ledger of the malicious vehicles within the network. In this paper, an innovative distributed framework is proposed for the identification and the tagging of malicious vehicles. This framework is based on Arabic license plate recognition using different image recognition algorithms and the identification of the vehicle as malicious or non-malicious propagate through the network, with higher accuracy in comparison to the other common plate recognition approaches. The details of both the vehicle communication framework and the image processing process are presented and the framework is validated through different implementations and discussion.