Implementation of a Terrain Radiometric Correction for UAVSAR
- Created: Thursday, 31 July 2014
A paper discusses a calibration method suited to correct variations of UAVSAR (unmanned aerial vehicle synthetic aperture radar) backscatter with topography. To use radar backscatter to estimate forest biomass on terrain with slopes, it is necessary to remove the effect of topography. The remaining signal should be related to biomass. The hybrid approach uses the radar line of sight to project an oversampled version of the Digital Elevation Model into radar coordinates for summation. Terrain topography has a major impact on the radar backscatter. Slopes facing the radar appear very bright while slopes facing away appear darker.
The calibration method presented here uses a combination of two types of corrections: homomorphic and heteromorphic. In the first case, the terrain is assumed homogeneous within the radar pixel to be corrected, while the second method assumes the terrain is rough.
The calibration process only uses the files delivered in a standard UAVSAR scene download. The method compensates for average residual aircraft motion while still accurately modeling the illuminated area over moderately steep terrain.
This work was done by Marc Simard and Scott Hensley of Caltech, and Bryan V. Riel of the University of Texas for NASA’s Jet Propulsion Laboratory.
This Brief includes a Technical Support Package (TSP).
Implementation of a Terrain Radiometric Correction for UAVSAR (reference NPO-47891) is currently available for download from the TSP library.
Please Login at the top of the page to download.