The Analysis of Hyperspectral Imagery and Testing the Wavelet Variogram

Abstract

This report contains one interim report, together with the most recent results. Part I describes the work done during a visit by Mr. Bosch of the Topographic Engineering Center, Alexandria. Part II describes the analysis of selected pixels from part of a SPOT image of Fort A. P. Hill chosen to represent several different types of ground cover. The classes have distinct characteristics evident in the variograms. Part III describes the analysis of hyperspectral data: the hydice imagery for Fort Hood and hymap imagery for Fort A. P. Hill for six different ground cover classes at each site. For Fort Hood the spectra vary considerable for the different ground cover classes and the variograms have certain distinctive features. The spectra for Fort A. P. Hill are less distinctive apart from one class, but the amplitudes of the periodic functions fitted to the variograms provide one means of discrimination. For both sites three principal components accounted for most of the variation. When the pixels were plotted in a projection of the first two components four clear groups were evident for Fort Hood and three for A. P. Hill. A non-hierarchical multivariate method of classification was also applied. Appendix I contains the GenStat program to fit periodic functions, Appendix II a paper submitted to the International Journal of Remote Sensing, and Appendix III a paper submitted to Mathematical Geology.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Apr 01, 2002
Accession Number
ADA400803

Entities

People

  • Margaret A. Oliver

Organizations

  • University of Reading

Tags

Communities of Interest

  • C4I
  • Materials and Manufacturing Processes
  • Sensors
  • Space

DTIC Thesaurus Topics

  • Amplitude
  • Birds
  • Data Sets
  • Drainage Basins
  • Earth Sciences
  • Economic Forecasting
  • Engineering
  • Geography
  • Hyperspectral Imagery
  • Information Science
  • Long Wavelengths
  • Periodic Functions
  • Remote Sensing
  • Soil Science
  • Statistics
  • Surveys
  • United States

Readers

  • Environmental Remediation and Restoration.
  • Image Processing and Computer Vision.
  • Regression Analysis.