A Service-Friendly Approach to Discover Traveling Companions Based on ANPR Data Stream

Abstract

Traveling companions are object groups that move together in a period of time. In this paper, we introduce a special kind of traffic data stream, which is called Automatic Number Plate Recognition (ANPR) data. In order to quickly identify traveling companions on ANPR data stream, this paper proposes an analysis algorithm named COINCIDENT. Owing to the privacy and security of ANPR data stream, the algorithm is encapsulated as a data stream service. The main contributions include: 1) we consummate our previous stream data service model and design a traveling companion discovery service on the model. 2) the proposed COINCIDENT algorithm can instantly and continuously discover traveling companions on the live ANPR data stream. Final experiments show the effectiveness and efficiency of our developed service.