Currently, agencies within the Department of Defense (DOD) and the private sector are trying to develop techniques capable of detecting the presence of small unmanned aerial vehicles (UAVs) at ranges on the order of 1–5 km. Often complicating this effort is the presence of small birds of similar dimensions that are frequently mistaken to be small UAVs when imaged using various methodologies, such as visible, short-wave infrared (SWIR) and thermal imaging.
Initially, research was limited to long-wave infrared (LWIR) thermal imaging that, in theory, should be effective for both day-and night-time operation. The goal was to record a preliminary set of calibrated radiometric and polarimetric images in which both small UAVs and birds were present in the scene. Targets (i.e., UAV and/or birds) within each image were segmented and regions of interest (ROI) were generated by tracing a circumference around the object. Pixel values were then averaged to yield both a radiometric value (watts/cm2-sr) and a corresponding polarimetric value (percentage linearly polarized) in order to quantify similarities and differences between “generic” bird and UAV signatures.
Although LWIR radiometric thermal signature capture and analysis techniques are very mature and well understood, thermal polarimetric methods and analysis are less developed. For the polarimetric analysis presented here, the well-accepted Stokes image methodology that effectively yields the percentage of the overall radiance emitted/reflected from the object that is linearly polarized will be used.
The most common approach for measuring the polarization state for emitted or reflected light is to measure the Stokes parameters: S0, S1, S2, and S3. For imaging applications these Stokes parameters are determined on a pixel-by-pixel basis in order to reconstruct a 2-D image. The Stokes parameters are determined by measuring the intensity of radiance that is projected through a polarizer/wave-plate pair that are oriented at various angles in order to measure the intensity of a particular state..
The Stokes images S0, S1, S2, and S3 are defined by Eqs. 1, 2, 3, 4:
S1 = I(0) – I(90) (w/sr-cm2), (1)
S2 = I(+45) – I(–45) (w/sr-cm2), (2)
S3 = I(R) – I(L) (w/sr-cm2), (3)
where I(0), I(90), I(+45), and I(–45) represent the measured radiant intensity of the linear states (measured relative to the vertical), at angles 0°, 90°, +45°, and –45°, respectively, and I(R) and I(L) represent right- and left-handed circularly polarized radiant states. The total radiance, S0, image (often thought of as just the intensity image) is defined as,
and the degree-of-total-polarization parameter/image is defined as,
However, for most applications that involve remote passive polarimetric imaging in the thermal IR, S3 is very small and rarely measurable and thus taken to be approximately zero. As a result, Eq. 5 is reduced to the more common degree-of- linear-polarization (DoLP) parameter/image and is defined as,
There are a variety of optical approaches/methods implemented in various polarimetric camera designs appropriate for polarization state filtering and analysis in the thermal IR. Examples include division-of-time approach, which uses a spinning-achromatic-retarder (SAR) arrangement, a division-of-amplitude (DoA) in which a polarized beam-splitting-plate is used to project different polarization states of the same scene onto one, two, or more focal-plane-arrays (FPA), and the division-of-focal-plane (DoFP) design in which micro-polarizers oriented at different angles are attached to individual pixels that make up the FPA. An example of an LWIR thermal image (left) and its corresponding polarimetric DoLP image (right) are shown in the accompanying image.
This work was done by Kristan P. Gurton for the Army Research Laboratory. For more information, download the Technical Support Package below. ARL-0238
This Brief includes a Technical Support Package (TSP).
Calibrated Long-Wave Infrared (LWIR) Thermal and Polarimetric Imagery of Small Unmanned Aerial Vehicles (UAVs) and Birds
(reference ARL-0238) is currently available for download from the TSP library.
Don't have an account? Sign up here.