Empregando Clustering para identificação de trajetórias de transporte de ilícitos por veículos

Abstract

Spatio-temporal data can be produced in an Intelligent Transport System (ITS) by License Plate Recognition devices, or LPR (License Plate Recognition), and are generated from the registration of vehicles through equipment, fixed or mobile , installed at certain points along routes along highways. These data are useful because they allow to reproduce the trajectories traveled by vehicles. This work seeks to identify trajectories used to transport illicit vehicles by land vehicles, presenting comparative results with significant trajectories, where the DBSCAN algorithm is able to identify the concentration of the most frequent trajectories formed by points of greater density.