SBIR-STTR Award

Automated tracking of moving vehicles is an enabling technology for monitoring the activities of specific threat targets, especially as related to the proliferation of weapons of mass destruction. Tracking techniques rely on repeated target observations over a period of time. However, traditional applications for spatial feature processing often fail when the moving target is obscured from view or is in close proximity to other similar vehicles. This project will focus on the use of visible and thermal hyperspectral imagery to measure a target¿s characteristic spectral signature and match the signature against future candidate target observations. The approach will enhance the performance of traditional high-resolution video trackers by reducing false alarms in ambiguous scenarios. Commercial Applications and other Benefits as described by the awardeeBeyond the application to national security, the technology should find use in law enforcement, border monitoring, and military applications, including (1) real-time detection, tracking, and identification of designated vehicles and dismounts, (2) wide-area persistent surveillance from remote airborne platforms, and (3) identification and mapping of temporally evolving threats such as chemical weapon releases.

Long-term continuous moving-target surveillance from airborne electro-optical sensors provides critical information for tactical awareness situations. Tracking civilian vehicles in urban environments is a challenging problem for existing systems, which generally rely on high-resolution video imagery to identify targets by their spatial characteristics. It is difficult for current spatial-based trackers to re-acquire target lock once the subject has been obscured from view for even moderate lengths of time. The goal of this effort is to demonstrate the exploitation of a vehicle target