Engineering topics
Publications and source records attributed to Dubois, P. C..
Intended as an overview aimed at potential users of remotely sensed spatial distributions and temporal variations of soil moisture, this paper begins with an introductory section on the fundamentals of radar imaging and associated attributes.
An unsupervised clustering algorithm is used to segment multifrequency polarimetric radar data from the NASA/JPL airborne SAR (synthetic aperture radar). Twenty-two parameters are evaluated for their discriminatory capability for each pixel of an image. A clustering analysis is then performed using different subsets of these parameters. This analysis relies on data taken as part of an intensive field experiment during the summer of 1988 in the vicinity of the Pisgah lava flow in the Mojave Desert in southern California. As part of the experiment, extensive ground truth was acquired, including dielectric constant and topography measurements. Segmentation results show good agreement with these measurements.
The results of inferring geologic parameters such as rms surface height, correlation length, and dielectric constant of rough surfaces by fitting observed polarization signatures with those predicted by the second order Rice model are discussed. The inferred results are compared to measured values of rms height and correlation length. The rms height values inferred are in good agreement with in situ measurements. The inferred correlation lengths generally do not agree with measured values. The results allow the separation of the effects of surface roughness and dielectric constant on the overall backscatter from rough surfaces.
The scattering matrix and Stokes matrix formats for multipolarization synthetic aperture radar (SAR) images are introduced. The effects of converting to the Stokes matrix format, without first doubling the sampling rate to allow for the conversion from a complex to an intensity format is quantified and discussed. It is shown that for most applications the Stokes matrix format is acceptable, since errors introduced by undersampling tend to average out when large numbers of pixels are averaged. For those applications requiring analysis of single image pixels, the scattering matrix format is recommended.