ALLEN, Tim, Geology Dept., Mailstop 2001, Keene State College, Keene NH 03435-2001, tallen@keene.edu.
Quantitative modal analysis of rocks is commonly done by point counting of thin sections. For very coarse-grained or heterogeneous rocks, however, the thin section no longer provides a representative sample of the rock. In some cases, it may be necessary to count modes at the outcrop. For a recent study of migmatite gneisses, the modes of leucosome (the light colored, granitic part), melanosomes (the dark colored, restitic part), and mesosomes (the remainder, of intermediate color and composition, possibly representing the original rock) were determined using two techniques. The first involved point counting with a wire mesh on the outcrop, and the second, classification of digital "grayscale" images obtained by scanning black-and-white photographs of the outcrop surfaces.
The classification involved determining the characteristic ranges (histograms) of brightness values of pixels within "training sites" for each of the three classes, selected by visual inspection of the image. These characteristic histograms were used to determine threshold brightness values that could be used to classify each pixel in the image as light (leucosome), intermediate (mesosome), or dark (melanosome). Once classified, the number of pixels in each class (the modes) can be easily determined. The digital classification was done using public domain software running on commonly available computer hardware. The results from this technique compared well to the outcrop point counting results, which suggests that digital image processing may be useful in other studies of coarse-grained or heterogeneous rocks.
1994 Geological Society of America Abstracts with Program 26(3): 2.