Spectral Assisted Moving Vehicle Tracking

The detection of activities related to the proliferation of weapons of mass destruction requires a capability of the automatic tracking of moving vehicles from an overhead platform. Current approaches rely on spatial-temporal characteristics coupled with moving object maps and tracking techniques. If the vehicle can be continually tracked, these techniques will work well. However, due to gaps in coverage, temporal tracking procedures can fall short of their goal, and additional characteristics that can uniquely identify a target vehicle must be considered. This project will investigate methods where the spectral observable of the moving vehicles can be used to improve the efficacy of tracking. Phase I focuses on understanding the phenomena associated with the spectral observables of vehicles for day/night longwave infrared (8 to 13 um) and day-only visible/near-infrared/shortwave-infrared (0.5 to 2.4 um) observation. Data will be collected in both spectral bands, and algorithms will be developed that can distinguish one vehicle from another by their spectral/thermal characteristics. Commercial Applications and other Benefits as described by the awardee: There is a direct public benefit from the implementation of a general area surveillance program. This has application to the monitoring of foreign war zones or the tracking of suspicious vehicles that could lead U.S. Forces to a hidden terrorist base. There are also applications to the problem of tracking small boat in harbors or coastal waters and to the application of the police of following a vehicle in traffic.