The mathematics of estimating overdetermined polarization parameters is worked out within the context of the inverse modeling of linearly polarized light, and as the primary new result the general solution is presented for estimators of the linear Stokes parameters from any number of measurements. The utility of the general solution is explored in several illustrative examples including the canonical case of two orthogonal pairs. In addition to the actual utility of these estimators in Stokes analysis, the pedagogical discussion illustrates many of the considerations involved in solving the ill-posed problem of overdetermined parameter estimation. Finally, suggestions are made for using a rapidly rotating polarizer for continuously updating polarization estimates.
A two meter inner diameter goniometer provides approximately 0.1° angular positioning precision for a series of spectral and polarimetric instruments to enable measurements of the directionality of polarized reflectance from soils in the laboratory, at 10° increments along the azimuth and zenith. Polarimetric imaging instruments to be mounted on the goniometer, with linear polarizers in rotators in front of each instrument, include broadband focal plane array imagers in the Visible band (Vis), Near InfraRed (NIR), Short Wave InfraRed (SWIR), and Long Wave InfraRed (LWIR) spectral bands, as well as a hyperspectral imager in the Vis through NIR. Two additional hyperspectral polarimetric imagers in the Vis through NIR, and SWIR, are to be mounted separately with angles measured by laser on the goniometer frame.
A key product of the global undisturbed/disturbed earth (GUIDE) program is the development of a soils database of broadband, hyperspectral, and polarized data. As a part of the GUIDE program, the U.S. Army Engineer Research and Development Center (ERDC) conducted a testing series involving a large variety of instrumentation at several sites at the Yuma Test Center (YTC) in fiscal year 2015 under the auspices of the Joint Improvised Explosive Device Defeat Organization (now the Joint Improvised-Threat Defeat Agency), generating approximately 17 terabytes of data. Most of this data, available through the ERDC, comprises hyperspectral polarimetric scientific data in the visible, near-infrared, shortwave infrared, and longwave infrared bands. As part of this testing series the performance of six handheld devices was characterized. We discuss the process of this data collection at YTC focusing on the polarimetric data, including the two handheld devices that relied on polarization for detection. Although some other polarization states discriminate soils better in some other wavelengths, for certain visible and near-infrared bands the Stokes S2 parameter provided the best discrimination.
Remote-sensing technology designed to exploit disturbed earth signatures has become extremely useful in the detection of disturbed soil in military areas of operation. Soil reflectance can be exploited for this purpose and is dependent on atmospheric conditions. An understanding of the in situ soil background is vital to any type of change detection. Researchers from the Engineering Research and Development Center (ERDC) conducted OCONUS soil spectral measurements at ten sites in Afghanistan from July to November, 2011. Sampling sites were chosen on the basis of geomorphic setting, surface-soil characteristics, and field-expedient conditions. Goniometric spectral measurements at these sites have provided high quality bi-directional reflectance data, and their analyses are presented in the context of threat recognition and discrimination. These data can also provide the basis for BDRF model validation. Most spectral data were acquired under ambient solar lighting, but other data were collected at night and under artificial illumination conditions. Bidirectional measurements of soil reflectance in the VIS/NIR and SWIR were taken using the University of Lethbridge Goniometer System (ULGS) at dawn, mid-day, dusk and after sunset with a light. Soil surface roughness and reflectance varied, depending on the presence of desert varnish and desert pavement at some sites. Sun angle and dust and smoke in the atmosphere impacted soil reflectance and noise in the SWIR part of the light spectrum, in particular. The presence of minerals such as calcium carbonate, gypsum, and oxidized iron in the subsurface directly impacted reflectance measurements in disturbed soil.