Remote sensing using hyperspectral and polarization images

We have developed relatively compact, lightweight, and programmable hyperspectral/polarization imaging systems using an acousto-optic tunable filter with different focal plane arrays to cover the spectral range from the visible to the long infrared wavelength region. Four separate imagers have been developed to cover this spectral region. In general, an AOTF is a polarization-sensitive tunable optical device. By combining it with a tunable retarder we can also collect the polarization signatures as well as the spectral signatures. We have such spectropolarimetric imagers in the visible-to-near infrared (VNIR, 0.4-1.0 mum) and short wave IR (SWIR, 0.9-1.7 mum) regions. A lot of remote sensing data using a VNIR spectropolarimetric imager have been analyzed using a commercial image processing software program (ENVI). In this paper we will discuss the result of this analysis. The VNIR imager was used to collect spectral and polarization data from various objects and backgrounds, both in the laboratory and in field tests. This imager uses a tellurium dioxide (TeO2) acousto-optic tunable filter (AOTF) and a liquid-crystal variable retardation (LCVR) plate with a charge coupled device (CCD) camera. The spectral images were collected from 0.45 to 1.0 mum with a 10 nm step, at two or four polarization settings for each spectral interval. We analyzed a portion of these data to assess the effectiveness of this system for foliage detection. Here we present our imager design, some results from our measurements and discuss the analysis results. Our results clearly show that compact, robust hyperspectral imaging systems with spectral and polarization detection capabilities will contribute significantly in a wide variety of future remote sensing applications.