Analysis of criminal behaviors for suspect vehicle detection

Abstract

The crime problems become critical issues for national security especially the security of border and intelligent transportation systems (ITSs). These affect the economy, investment, tourism, and society. As a result, the automatic suspect vehicle detection emerges as one of effective tools to tackle the problems. However, the traditional process normally uses criminal vehicle data in blacklist comparing with vehicle data gathering from various sensors. This comparison is not effective and accurate that might be from not up-to-date data in the blacklist. Sometimes the blacklist is not available. This paper proposes the criminal behavior analysis method to detect suspect vehicles that are potentially involved in criminal activity. It must not rely on the blacklist. The analysis is conditional on journey path and the involvement of criminal activities. In additional, public officials believe that the suspect vehicle will choose the journey path without a checkpoint. Therefore, we used the journey path analysis techniques together with the association rule mining to analyze such criminal behavior. From extensive experiments, the results show that the proposed method can increase the suspect detection accuracy rate 17.24% beyond the traditional counterpart.