You are here

Three-dimensional Context and Feature Aided Multitarget Tracking

Award Information
Agency: Department of Defense
Branch: Air Force
Contract: FA8650-09-M-1595
Agency Tracking Number: F083-172-0096
Amount: $99,952.00
Phase: Phase I
Program: SBIR
Solicitation Topic Code: AF083-172
Solicitation Number: 2008.3
Timeline
Solicitation Year: 2008
Award Year: 2009
Award Start Date (Proposal Award Date): 2009-01-23
Award End Date (Contract End Date): 2009-10-31
Small Business Information
4850 Hahns Peak Drive Suite 200
Loveland, CO 80538
United States
DUNS: 956324362
HUBZone Owned: No
Woman Owned: No
Socially and Economically Disadvantaged: No
Principal Investigator
 Juan Vasquez
 Program Director
 (937) 427-9725
 juan.vasquez@numerica.us
Business Contact
 Stephanie Mueller
Title: Contract Manager
Phone: (970) 461-2422
Email: stephanie.mueller@numerica.us
Research Institution
N/A
Abstract

Dismount discrimination is often based on static EO or IR target attributes such as height, length, and color. Various algorithms such as gait estimation have been developed to distinguish dismounts from other moving objects. Techniques that build high-fidelity 3D models of dismounts are fragile when applied to real imagery (i.e., video mounted on an unmanned aerial vehicle (UAV) with limited communication bandwidth and operating in windy conditions). A novel approach is to exploit the information derived from the track state being provided by a tracker to enhance existing methods for gait estimation and 3D target modeling. Specific attention will be given to managing lower resolution EO/IR data and combining the strengths of recently developed concepts in order to provide robust methods for dismount discrimination and feature aided tracking. In addition, micro-doppler radar features will be exploited to provide dismount gait estimation, which leads to the benefit of a layered sensing approach. The micro-doppler radar features are highly dependent on target orientation to the sensor. Knowledge of this orientation provided by other layers in the sensing network will significantly enhance the radar-based discrimination and classification. The layered sensing paradigm provides opportunities for data fusion at the measurement or track level. BENEFIT: The specific problem of three-dimensional multi-intelligence surveillance systems is addressed. The developmental sections of the software can be translated into a real-time system for direct use by the military and civilian reconnaissance communities. Specific DoD programs include instantiations of several UAV sensor platforms, the recently deployed Angel Fire system, and the evolving ARGUS-IS persistent video platform. UAV video platforms are becoming increasing prolific and provide new opportunities to employ layered sensing architectures. Demonstration of Numerica’s capabilities in the layered sensing problem domain will be directly marketable to civil application areas such as include border patrol and urban surveillance. The recent increase in aerial systems capable of providing persistent radar or video will be appropriate for homeland defense applications that require wide area sensor coverage along with high-confidence tracking solutions. The tracking technology developed under this SBIR will be extensible to private surveillance systems without the need to modify existing hardware systems. As such, a licensable version of a matured software suite will be realizable. Numerica has on-going relationship with Boeing, Lockheed Martin, Raytheon, and various other DoD contractors with programs related to surveillance systems.

* Information listed above is at the time of submission. *

US Flag An Official Website of the United States Government