Geology Reference
In-Depth Information
will have slightly higher gravitational accelerations than
areas of low-density materials, when differences in local
topography are taken into account. In planetary explora-
tion, mapping gravity variations, as was done for the
Moon, provides insight into subsurface structure and the
possible relation to surface features.
Magnetic
fields vary in space and time on planets. On
Earth, measuring the orientations of the
fields.
“
locked
”
in
rocks helped build the case for the theory of plate tectonics
and showed that the magnetic poles have shifted with
respect to the spin axis. Not all planets have magnetic
fields today, but magnetometers
flown on spacecraft have
revealed the presence of remnant magnetic
fields. For
example, the Mars Global Surveyor spacecraft recorded a
remnant
field in the oldest rocks exposed on the surface.
Its absence in the younger terrains suggests that there was
a fundamental change in the interior of Mars early in its
history, thus demonstrating that geophysical measure-
ments combined with surface geology provide powerful
tools for deciphering planetary histories.
2.7 Image processing
Digital
“
snapshot
”
cameras and computers typically have
built-in routines to manipulate images, such as
“
red-eye
”
removal. Such digital image processing has become so
commonplace that many people do not understand what is
involved. In this section, some of the principles and prac-
tices of planetary image processing are outlined.
The basic units of digital images are pixels (an abbrevia-
tion for
), which can be arrayed in hori-
zontal lines and vertical rows tomake a picture. The amount
of light received by the detector for each pixel is encoded by
its brightness level. In 8-bit encodement (2 raised to the 8th
power), 256 shades of gray can be assigned, with each level
referring to a speci
c
“
picture elements
”
Figure 2.15. Images of Mars to illustrate some common image
processing techniques: (a) calibrated image showing a histogram
(bottom of image) of DN levels within a relatively narrow range
centered at about 134; (b) a
image inwhich the DN levels
are spread over a wider range than in (a) and shifted toward a darker
(lower DN) level; (c) a low-pass
“
stretched
”
“
DN
”
(digital number),inwhicha
DN of 0 is black (no light received) and a DN of 255 is a
perfectly white level. These are typically shown on an
image by a DN histogram that gives the distribution of the
various levels of gray in an image (
Fig. 2.15
). Because the
human eye can discriminate only about 30 shades of gray,
this means that digital images potentially contain much
more information than can be visually detected. Various
algorithms can be applied to extract this information after
calibration of the image.
Calibrations. Digital image detectors are not uniform in
their response to light. Within any given array, some detec-
tor pixels will be more sensitive than others. Consequently,
filter image resulting in a smoother
appearance; (d) a high-pass
filter image in which differences in
brightness are enhanced; and (e) a sharpened image in which
boundaries among DN changes are emphasized (part of Mars Orbiter
Camera image mc27
-
256).
after the camera has been assembled, but before it goes into
use, it must be calibrated to take these differences into
account. In its simplest form, an image of a
”
which consists of a perfectly uniform, perfectly illuminated
surface is taken by the camera. Each pixel is then adjusted
by adding or subtracting DNs until all the pixels have the
same value. This mapping of DNs is then standardized for
“
“flat field,”
