Real-time wide area tracking: hardware and software infrastructure

In recent years, large number of cameras have been installed in freeway and road environments. While the use of some of these cameras is being automated through computer vision, few computer vision systems allow for true wide-area large scaled automation. This paper describes a distributed computing system capable of managing an arbitrarily large sensor network using only common computing and networking platforms. The architecture is capable of handling many common computer vision tasks, as well as the inter-sensor communication necessary for developing new algorithms which employ data from multiple sensors. This system is tested with an algorithm which tracks moving objects through a prototype camera network with non-overlapping fields or view in a college campus environment This algorithm allows the system to maintain the identity of a tracked objects as it leaves and enters the fields of view of individual sensors. Such an algorithm is necessary for applications which require tracking objects over large distances or over long periods of time in an environment without complete sensor coverage.