ISPRS Commission III, Vol.34, Part 3A „Photogrammetric Computer Vision“, Graz, 2002 
The assessment of the results was performed by comparison of 
the results of the automatic interpretation, as described in this 
paper, and a manually created biotope mapping for the last 
epoch 1998. Two points were included into the comparison: the 
segment borders and the classification of the segments. The 
segment borders were in most cases similar, as well as the 
classes. Following differences could be found: Some areas, 
which were manually classified as “forest”, were assigned to 
“area of regeneration birch state” by the automatic system. 
Also, the opposite case could be found. The reason is that there 
is no sharp separation between transition of these two classes. 
The question, when the class “area of regeneration birch state” 
ends and when the class “forest” begins is more or less 
subjective. 
In some other areas the automatic system classified “area of 
regeneration” as “area of degeneration”. The reason is that both 
classes look very similar. The assignment to one of these 
classes can only be done by using temporal history of the 
segments (see above). For the misclassified cases the automatic 
system did not have enough temporal information for the 
epochs before 1975. The use of images from epochs before 
1975 would probably lead to correct results in those segments. 
8. CONCLUSION 
A system for knowledge based multitemporal interpretation of 
aerial images was presented. The explicit knowledge 
representation allows an easy integration of expert knowledge 
into the system. For interpretation of vegetation areas the 
concept of manual interpretation by using interpretation keys 
was transformed into an automatic interpretation system by 
using feature analysis operators. For interpretation of temporal 
changes an approach was presented, which discretely describes 
temporal conditions of regions, and which transfers the most 
probable temporal changes of the given conditions as temporal 
knowledge into a state transition diagram, then using it for 
multitemporal interpretation. 
Based on these approaches a procedure for automatic 
multitemporal interpretation of industrially used moorland was 
successfully developed. Proceeding from an initial 
segmentation based on Geo-Data resegmentation and 
interpretation of the segments is carried out for each 
investigated epoch. By using temporal knowledge it is possible 
to separate moor classes, which can only be detected in 
temporal order. The application of temporal knowledge and 
structural features enables the exclusive use of greyscale images 
for interpretation of vegetation areas. The results show that the 
presented procedure is suitable for multitemporal interpretation 
of moorland, and that it is able to distinguish additional moor 
classes compared to the approaches used so far. It is further 
applicable for a more robust multitemporal interpretation, and 
does not depend on colour images. 
In some parts this work contains potential for improvements. 
Although the feature analysis operators are designed to work 
with a minimum of parameters, their automatic adaption to the 
used images would improve the system's level of automation. 
Further parts are resegmentation and probabilities of 
multitemporal interpretation. Additionally, the suitability of the 
used prior knowledge should be verified for other moor areas 
and other applications. 
9. REFERENCES 
Eigner, J,  Schmatzle, E, 1991. Handbuch des 
Hochmoorschutzes - Bedeutung, Pflege, Entwicklung. Kilda- 
Verlag, Greven, 158 p. 
Forstner, W., Liedtke, C.-E., Biickner, J. (Eds.), 1999. 
Workshop on Semantic Modelling for the Acquisition of 
Topographic Information from Images and Maps (SMATI'99). 
Proceedings, Bonn, 227 p. 
Growe, S., 2001. Wissensbasierte Interpretation 
multitemporaler Luftbilder. Dissertation, Universität Hannover, 
Fortschritt-Berichte VDI, Reihe 10, Nr. 656, VDI-Verlag, 
Düsseldorf, 144 p. 
Heipke, C., Pakzad, K., Straub, B.-M., 2000. Image Analysis 
for GIS Data Acquisition. Photogrammetric Record, 16(96), 
pp. 963-985. 
Liedtke, C.-E., Bückner, J., Grau, O., Growe, S., and Tónjes, 
R. 1997. AIDA: A system for the knowledge based 
interpretation of remote sensing data. 3rd. Int. Airborne Remote 
Sensing Conference and Exhibition, Vol. II: pp. 313-320. 
Lunetta, R. S., Elvidge, C. D. (Editors), 1999. Remote Sensing 
Change Detection — Environmental Monitoring Methods and 
Applications. Taylor & Francis, London, 318 p. 
Mayer, H., 1998. Automatische Objektextraktion aus digitalen 
Luftbildern. Deutsche Geod. Kommission, Reihe C, Nr. 494, 
132 p. 
Niemann, H., Sagerer, G., Schróder, S. and Kummert, F., 1990. 
ERNEST: a semantic network system for pattern understanding. 
IEEE Transactions on Pattern Analysis and Machine 
Intelligence, 12(9), pp. 883-905. 
Pakzad, K., 2001. Wissensbasierte Interpretation von 
Vegetationsflächen aus multitemporalen Fernerkundungsdaten. 
DGK, Reihe C, Dissertationen, Nr. 543, München, 104 p. 
Pakzad, K., Growe, S., Heipke, C., Liedtke, C.-E., 2001. 
Multitemporale Luftbildinterpretation: Strategie und 
Anwendung. Künstliche Intelligenz, (15) 4, pp. 10-16. 
Peled, A., Haj-Yehia, B., 1998. Toward automatic updating of 
the Israeli National GIS - Phase II. International Archives of 
Photogrammetry and Remote Sensing, Vol. 32, Part 4, Stuttgart, 
pp. 467. 
Tónjes, R., 1999. Wissensbasierte Interpretation und 3D- 
Rekonstruktion von  Landschaftsszenen aus  Luftbildern. 
Dissertation, Universität Hannover, Fortschritt-Berichte VDI, 
Reihe 10, Nr. 575, VDI-Verlag, Düsseldorf, 117 p. 
Von Drachenfels, O., 1994. Kartierschlüssel für Biotoptypen in 
Niedersachsen. Naturschutz und  Landschaftspflege in 
Niedersachsen, Niedersüchsisches Landesamt für Okologie, 
192 p. 
Weismiller, R. A., Kristoof, S. J., Scholz, D. K., Anuta, P. E., 
Momen, S. A, 1977. Change Detection in Coastal Zone 
Environments. Photogrammetric Engineering and Remote 
Sensing, 43, pp. 1533-1539. 
A - 239