Overhead Persistent Infrared Tracking

ABSTRACT:The GATER processing framework was originally developed to support full motion video (FMV) data sources. The software has been very successful in this domain, and is now relatively mature after 3 years of development. Extensions are required to the GATER framework when targeting OPIR data sources. Challenges unique to OPIR data render the processing problem fundamentally different than that for FMV data. These differences include low-level sensor differences, differences in the platforms typically used to acquire OPIR data, and differences in the exploitable scene contents. Etegent has developed a variety of algorithms that address concerns pertinent to the OPIR tracking problem, such as expectation maximization principal component background suppression, kernel based super-resolution dejittering. These tools as well as a few novel approaches have been identified as candidates to enhance the robustness of OPIR GATER for a variety of data products. Additionally, not all approaches are suitable for all OPIR sensors or surveilled scenes, which implicitly require the proposed modeling of the system performance in order to optimize the system as a function of the operating conditions.BENEFIT:The research conducted under this Phase I effort will lay the groundwork for the fuller development of a suite of efficient, scalable tracking algorithms within the GATER framework under a Phase II effort, and for evaluation of tracker efficiency under differing operating conditions. Several of the algorithms being considered for development under this Phase I effort will also have more specific, immediate applicability outside of the context of this project as well.