method
.16
9
.18
.10
AS
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B3. Istanbul 2004
Figure 2A and 2B summarise the result of the analysis
in table 5. Line B refers to standard errors in x while
line C to that of y. From tables 4 and 4 and the graph
above there is no significant difference between the
accuracy that can be achieved by the two methods. In
both cases, particularly between exposure time of 1,5
and 2.5 seconds the accuracy ranges between 0.1 and
0.2 pixel. Considering the fact that the accuracy above
are for one multiple image, for four multiple images it
is possible to achieve an accuracy of 0.1 pixel or higher.
This corresponds to accuracy better than 0.3 arcsecond
for imaging systems of focal length greater than or
equal to 80-cm. and pixel size of 10um. The effect of
the apparent cloudy weather on the standard errors
shown in table 4 an 5 and the erratic behaviour of the
graph above indicates that one of the main limiting
factors of the centroid determination of star objects is
the prevailing weather condition rather than the method
used. However it appears that moment analysis method
is more prone to bad weather.
5. Conclusion
The above experiment, analysis and results show that
an accuracy of 0.1 pixel can be achieved and that there
is no significant difference in the accuracy that can be
achieved between the two main methods of centroid
estimation, viz., moment analysis and PSF fitting. It
also confirms the theoretical findings mentioned in
section [2.2]. This indicates that it is possible to obtain
image measurement accuracies better than the +0.3
arcsecond required for astro-geodetic determinations
using portable telescopes, i.e., focal lengths between 80
and 150 cm The analysis also showed that centroid
estimates deteriorates with bad or cloudy weather more
seriously in the case of moment analysis. Furthermore
during the centroid measurements it was found that the
PSF fitting method is more adaptive to automation.
References
1. Bove, V.M. Jr. (1993); Entropy based depth
of focus. Journal of Optical Society of
America Volume 10 No. 10
2. Buil, C. (1991); CCD Astronomy: CCD
Astronomy, construction and use of an
Astronomical CCD Camera. William-Bell
Inc. Virginia
5. Chubunichev, A.; Algorithms of digital
target location and their investigations. ISPRS
Journal Vol. XXIX Commission V
4. Eisfeller B., G.W.Hein; (1994);
Astrogeodetic levelling with an Integrated
DGPS/CCD star camera system. Proceedings
of the International Symposium on
Navigation (KISS 94) Calgary Canada.
s. Horn, B. K. P. (1986); Robot vision. The
MIT Press, Massachusetts.
617
10.
11.
13.
14.
Gonzalez, R.C., P. Wintz (1987); Digital
image ^ processing. Addison Wesley
Publishing Co.
Jain, A. K. (1989); Fundamentals of image
processing. Prentice Hall Inc., New Jersey
Kovalevsky, J. (1995); Modern Astronomy.
Springer Verlag, Berlin Heidelberg
Longhurst, R. S. (1967); Geometrical and
Physical Optics. Longmans, London.
Ploner, Martin (1996); CCD-Astronomie von
objekten des geostatiären ringes.
Geowissenschaftliche Mitteilungen Heft 46.
Technical University Vienna.
Raab, H. (1996): Astrometrica software
version 3.1
Schildknecht, T. (1994); Optical astronomy
of fast moving objects using CCD detectors.
Swiss Geodetic Commission.
Teuber, J. (1993); Digital image processing.
Prentice Hall International
