Geology Reference
In-Depth Information
Fig. 6.11.
Examples from 'Visual comparison charts' developed by Matthew et al. (1991) for studying ceramic materials.
In
contrast to the Baccelle and Bosellini charts, these charts also compare differences in grain sizes and sorting. They consist of
24 pairs of charts showing black grains on a white background or vice versa. The charts show two major groups - images with
rounded grains and images with mixed rounded/polygonal grains. The grains range in size from 0.5 to 10 mm. The images
display mixtures of different-sized grains, shown at percentages of 10, 25 and 50% and within different grain-size ranges. For
additional information and comment see Appendix (Comparison charts).
6.2.1.3 Image Analysis
differences in gray levels in an image of thin sections
under plane light or cathodoluminescence.
Image acquisition of thin sections studied under the
microscope usually takes place via a video scanner. Im-
age digitization then converts the analog TV signal into
a digital form. Further treatment includes image seg-
mentation with the purpose of enhancing the image and
discriminating between the phases that are to be mea-
sured in the sample. Limestones and dolomites need
specific preparation (e.g. etching, staining) in order to
enhance the contrasts between different grain catego-
ries and to count frequencies (Harvey and Steinmetz
1971). Grain types, matrix, and cement in carbonates
can be made visible by using methods developed for
aerial image interpretation (Conley and Davis 1973).
Image analysis is important for quantifying textural cri-
teria and the composition of rocks at various scales.
Digital image analysis aids in determining the fre-
quency, shape, rounding, and size of grains and pores
and in determining statistical parameters that can be
plotted in graphs. The data are derived from micropho-
tographs, field photographs, images of thin sections or
peels studied with digital video camera, scanner or digi-
tal SEM. The back-scattered electron imaging mode of
the SEM provides fully automatized compositional and
morphological analyses of sandstones (Ehrlich et al.
1984; Krinsley et al. 1998) and carbonate rocks (Rogen
et al. 2001). Several computer-assisted systems are com-
mercially available (e.g. Kontron IBAS, GIPSY,
PETROG or DIANA 2001). Most image analysis soft-
ware packages allow grains (Van den Berg et al. 2002),
pores, and cements to be differentiated by recognizing
Potential of image analysis for carbonate studies:
The method allows a wealth of different characteris-
tics of carbonate rocks including color and weathering
