A study has investigated the ability of ratioed information from Landsat 2 Multispectral imagery, to discriminate geological features through the application of photometric processing techniques. United States Geologic:.1 survey maps corresponding to the areas investigated have been used in correlation with ratioed imagery to identify predict and confirm geological features on the basis of resulting densitometric information. A multivariate, discriminant analysis of rock type, exposure data has enabled the fabrication of the final, optimized ratio image. The optimisation process used to predict geological types involved a priority structure that found the spectral bands with the highest discriminatory power and the least correlation and then predicted contrast values for the positive and negative masks. Lab test imagery of several rock samples provided impetus for further investigation from the chosen orbital imaging system, the Landsat Satellite series.
Library of Congress Subject Headings
Marsh, Donald, "Photometric processing and interpretation of ratioed imagery by multivariate discriminant analysis for separation of geological types" (1984). Thesis. Rochester Institute of Technology. Accessed from
RIT – Main Campus