Every day millions of people drive along urban roads and highways featuring driving own behaviors. When considering the highway network, great efforts have been employed to monitoring such large number of users considering traffic code infringements, accidents detection or traffic congestion estimation. Such information are useful to reveal hidden drivers’ habits: in this study an analysis on traffic data has been developed to highlight suspicious events of highway users exploiting a dataset of the Italian Traffic Police. The analysis employed unsupervised clustering techniques and a series of filters on possible routes useful to isolate suspicious car stops and meetings on service areas.
An unsupervised behavioral analysis of highway traffic flow for security applications
Balducci F.;Calvano G.;Impedovo D.;Pirlo G.
2020-01-01
Abstract
Every day millions of people drive along urban roads and highways featuring driving own behaviors. When considering the highway network, great efforts have been employed to monitoring such large number of users considering traffic code infringements, accidents detection or traffic congestion estimation. Such information are useful to reveal hidden drivers’ habits: in this study an analysis on traffic data has been developed to highlight suspicious events of highway users exploiting a dataset of the Italian Traffic Police. The analysis employed unsupervised clustering techniques and a series of filters on possible routes useful to isolate suspicious car stops and meetings on service areas.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
