Amplitude Distributions for Seaset SAR and Skylab Data with Statistics for Classification and Correlation

Publisher Website
Google Scholar
Abstract
Seasat SAR and Skylab scenes of Nantucket Island, Mass., have been mapped to the same scale and orientation for a comparison of their reflectance distributions and for a classification of terrain. Rayleigh and Poisson statistics are employed in random Monte Carlo computer simulations to model the intensity distributions of the reflectance data. Two-dimensional autocorrelation functions for SAR and optical data and an amplitude normalized cross correlation of the multisensor scenes are computed with fast Fourier transform methods. Both positive and negative correlations are observed in the cross correlation, depending on the tilt modulation and texture of scene features. Six classes of terrain are identified using the Skylab green optical bandwidth, which is particularly sensitive to subsurface topography, and the Seasat L-band synthetic aperture radar, which is particularly Insensitive to smooth horizontal surfaces.

This publication has 3 references indexed in Scilit: