15
58/
|
| S
05
85
uoZo
)etter
and
e not
tems.
anual
igital
rs are
"user,
1atic
ntrol
hundred thousands of points. Both systems provide certain
indications of the matching quality and means to edit the data.
DPW 770 stores in a file a quality code from 0 to 100. Points with
code less than 33 are considered unreliable and are mostly
interpolated from other good points (see Figure 5). In a test all
these points (11.296) were deleted and the rest was interpolated in
a 5 m regular grid of the manual measurements for accuracy
analysis (see Tables 1 and 2). Only 39% of the deleted points (i.e.
ca. 4% of all points) were in reality blunders (error > 3 m). The
resulting RMS was much lower and the big blunders were
severely reduced. Thus, operationally all low quality code points
could be deleted and manual measurements could be made to fill-
in the resulting DTM gaps. Still, the remaining blunders (» 3 m)
are 4% (ca. 15,000 points) and not indicated by the quality code.
VirtuoZo divides the points in three quality classes that are
displayed on the screen with different colours. The worst class
includes points that have been interpolated. Again blunders are
included in the "reliable" classes and additionally this quality
code is not saved in any file for user access. So, the quality
measures provided by the systems must be made more reliable
and additional diagnostic tools should be used. However, DPW
led after deletion of the low quality code points to 2.5% less
blunders than VirtuoZo, which means less editing time or higher
DTM accuracy, if no manual editing is performed.
Both systems provide tools to indicate errors (contours, 3D
views, overlay of contours on orthoimages or in the stereomodel)
and means to edit the data pointwise or regionwise, but since the
blunders are not reliably highlighted, the user must control the
whole data set. The editing tools should, at least partially, used
before matching (e.g. exclusion of textureless areas, water bodies
etc.). Thus, time is saved and the given information can be used
in matching. Nevertheless, automatic DTM generation does have
certain advantages over manual measurements like processing
speed (over 100 points per sec) and the possibility to derive a
very dense DTM. This is significant because it leads to a much
more accurate terrain representation (implicit measurement of
breaklines, form lines etc.).
5. COMPARISON BETWEEN DIGITAL AND
ANALYTICAL SYSTEMS - CONCLUSIONS
Compared to analytical plotters digital photogrammetric systems
show following characteristics. The price of digital
photogrammetric systems is similar to much lower that the one of
the analytical plotters. They also offer additional functionality
like geometric sensor models for satellite imagery, image
rectification, orthoimage and orthoimage map generation,
mosaicking, 3D perspective views, animation and flyovers,
image processing, support of many input/output raster and vector
data formats etc. The cheaper systems however do not have all
these functions. VirtuoZo does not have for example
triangulation, mapping software (except simple feature
digitisation with an attribute code using an extra package),
orthoimage map generation (except of overlay of contours on
orthoimages), animation and flyovers, and image rectification.
Deficiencies that were observed during this project include the
following:
* The systems are not open and flexible enough. The content
and format of some files is unknown (e.g. metadata image
files with orientation etc. in DPW). Intermediate results like
109
pixel and photo coordinates can not be accessed. The user is
bound to the standard streamlined processing flow, so limited
operations (like import of a foreign orthoimage and creation
of an orthoimage map in DPW) are not possible or very
cumbersome.
VirtuoZo does not provide some common input/output
formats. Both systems input uncommon image formats (raw,
VITEC) which leads to an increase of required disk space
and processing delays.
* The documentation, especially of VirtuoZo, is poor and
explains very little on what exactly and how it is done.
Fully fledged systems like the DPW 770 are quite complicated to
handle and require significant training, whereby VirtuoZo is
easy-to-use and its user interface quite intuitive. The size of
digital data is still a problem. Large images can only be processed
a few at a time or not at all. Many processes appear as a blackbox
to the user, case which is not always desirable especially in
education and research. Digital photogrammetric systems have to
a large extent transferred the working mode of the analytical
plotters in the digital domain. Thus, for example they concentrate
on processing of stereo images without exploiting the existing
possibility of simultaneous processing of multiple images in
order to improve accuracy and reliability of the results
(especially for DTM generation and automated feature
extraction). Functionality should be extended (semi-automated
feature extraction, monoplotting, structuring of measurements,
import and integration of other data, data/metadata storage and
management, GIS functionality for data analysis). A comparison
between DPW 770 and VirtuoZo shows that both systems have
their strong and weak points. DPW 770 has more functions, but is
also more expensive and complicated to use. VirtuoZo on the
other hand offers the advantage of user-friendliness and low
price. It can certainly be used for DTM and orthoimage
generation, but it is not suitable for aerial triangulation or
mapping. Even with their present weaknesses digital
photogrammetric systems offer significant advantages in
comparison to analytical plotters and can be fully employed in
research and professional practise. The expected improvement of
their performance, especially in the algorithms and the user
interface, will expand their use even further.
Acknowledgements
The authors would like to thank A. Kääb and H. Bósch,
Glaciology, ETHZ for proposing the glacier project, supplying
data and helping in the recognition of the control points, as well
as the students A. Bleisch and C. Räuftlin for their engaged work
on this project within their “Praktikum” in photogrammetry.
References
Miller, S.B., Thiede, J. E., Walker, A. S., 1992. A Line of High
Performance Digital Photogrammetric Workstations - The
Synergy of General Dynamics, Helava Associates, and Leica.
In: International Archives of Photogrammetry and Remote
Sensing, Vol. 29, Part B2, pp. 87 - 94.
Zhang, J., Zhang, Z., Wu, X., Wang, Z., Qiu, T., Chao, H., 1994.
A Photogrammetric Workstation from WTUSM. In:
International Archives of Photogrammetry and Remote
Sensing, Vol. 30, Part 3/2, pp. 939 - 944.
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B4. Vienna 1996
