anag 3D Y2.0 - [3k Ansicht (48%]]
Datei Bearbeiten Dptionen
se >
Zwischenablage: i £ Aheinz\China-11-2001\5tereobilder Grube 2\g2-m-2.jpa
_ 0. hd
2160x1440
Figure 3. Generation of an anaglyph image using CRANAG 3D
adjustment of the images, the new stereo image is cropped and
saved as a new image file.
Different tools are available for this task. The one used for this
project is CRANAG 3D by Alex Leuthold (cf. fig. 3,
www .leual.ch/anaglyph/). It is free for noncommercial use and
supports different types of anaglyphs with different color
channel combinations and even shutter glasses. The adjustment
of the two images can be achieved by moving and/or rotating
the images. The result can always be viewed in a 3D-window. It
is recommended to adjust the images in a way that the
parallaxes vanish for points or areas that can be observed well
in the main plane. If the images are taken according to the
statements above, it should be sufficient to first move the
images, control the points, do a rotation if necessary, and in a
second step refine these adjustments. Another tool (freeware) is
Z-Anaglyph by Georges Rosset (z-graphix.com/anaglyph/
zanag_en.htm). It supports red-cyan anaglyphs only and has no
support for rotating images. The images must be taken quite
exactly or rotated with other tools.
The manipulation of the images can also be performed with
standard image processing software like e.g. Corel PHOTO-
PAINT® that supports rotation of images, the separation and
combination of color channels and the manipulation, e.g. copy
and paste, of images or parts of images. But the above
mentioned tools are especially designed for this task and
therefore very helpful, as they can speed up the process of
generating anaglyph images very much. As mentioned above, it
is necessary to keep the geometric conditions for image capture
in mind if you want to use these tools successfully and generate
impressive and suitable stereo images.
The anaglyph images can be printed on any inkjet color printer.
The paper should be of good quality. It is helpful to mark on the
prints from which direction this image is to be viewed with
which color combination of glasses. Glasses with the
corresponding color combinations are requisite for stereo
viewing and can be purchased for little money.
DATA PROCESSING AFTER THE FIELD CAMPAIGN
Image mosaicking. Back in the office after the end of the field
campaign the rectified images of the single fields were
combined using image mosaicking. Mosaicking is the process
of joining images together to form a larger image. The result is
a single image map of each location in a scale of 1 : 5. This map
is helpful for overview purposes and because of the larger scale
of the images. On the other hand, it is not essential as the
information itself has already been provided with the single
rectified images.
There are two possible problems in the mosaicking. Firstly,
there are differences in brightness, contrast and colors of the
single images. These can be adjusted approximately selecting
one image as a reference and proceeding step by step with the
surrounding images. Secondly, there will be geometric
differences between the different images caused by the fact that
the rectified images are no real orthophotos. This cannot easily
be corrected and will thus lead to certain cuts or unsharpnesses
in the overlapping areas depending on the procedure used for
the mosaicking. There are different methods to perform this
mosaicking. The simple one is to combine the single rectified
images with standard image processing software. In the
overlapping areas, cutlines can be generated which define the
image from which the information is used for the final mosaic.
Another way is to use special software tools for this purpose
—382-
D
positic
positic
from t
be ref:
Photo
compl
camer
points
points
plotter
measu
the co
mm w
image
accura
image
in con
take tk
After
Zeiss
Micro
of the
every
