Building Social Networks in Persistent Video Surveillance

Abstract

Social networks are a beneficial analysis tool in counterterrorism and counterinsurgent activities. The difficulty lies in the amount of time and resources it takes to construct a social network. By exploiting existing 24-hour overhead persistent video, we can build a social network from vehicle to vehicle and vehicle to building interactions. This paper demonstrates building a social network from vehicle tracks based on their interactions in an urban environment. From this social network we can see relationships among actors and their locations of interest. This information provides additional intelligence about terrorist activities to exploit them.