POLARIS SENSOR TECHNOLOGIES INC — Department of Defense SBIR Phase I: A22-011

Description

Detection, Recognition, and Identification of targets with infrared cameras is often limited by Fixed Pattern Noise (FPN).  This is especially true for polarimetric imagers that subtract images of orthogonal polarization to observe small differences that are typically less than 5%.  FPN is mitigated with Non-uniformity Corrections (NUC) algorithms and procedures.  NUC procedures can be grouped into those requiring images of a uniform external source such as a blackbody source, and those that can use the scene itself for correction.  The latter is referred to as a Scene Based NUC (SBNUC) and will be the focus of the proposed work.  Polaris Sensor Technologies (PST) will team with Dr. Brad Ratliff who has worked on SBNUC algorithms for several years and did his PhD work on this Topic.  The team combining PST with intimate knowledge of sensor physics and Dr. Ratliff with a tremendous background in SBNUC technology is an ideal team for the proposed work.   At the conclusion of the Phase I effort, we anticipate a real-time implementable SBNUC algorithm catered to LWIR polarimetric sensors that will significantly reduce residual FPN and ultimately improve the quality of estimated Stokes imagery for a range of scenarios of interest to the sponsor.