owth ver- 
‚tives. 
ld belong 
level and 
plex pro- 
othesized 
othesized 
gorithms 
e is that 
dge level 
store the 
isting or 
rosses in 
gmented 
Yee built 
Learning 
| estima- 
e, we do 
research 
of spa- 
generic 
it Com- 
udy the 
yze the 
Gram- 
nent of 
Univer- 
; 3-D 
Trans. 
|. topo- 
Com- 
onsen- 
Image 
1981, 
tische 
und - 
d. 
nages. 
91. 
tions, 
ature 
nding 
ction 
ln 
rdam 
e on 
ress, 
LAYTON M. (1987): A process-grammar for representing shape. 
Proc. Workshop ”Spatial Reasoning and Multi-Sensor Fusion”, 
1987, pp. 148-157. 
LECLERC Y. G. (1989): Constructing Simple Stable Descriptions 
for Image Partitioning. IJCV 3, 1989, pp. 73-102. 
MiNskKY M. (1986): Society of Mind. New York: Simon and Schus- 
ter. 
MiNSKY M. (1990): Thoughts about Artificial Intelligence. In 
[KURZWEIL 1990]. 
PAN H.P. (1990): A Spatial Structure Theory in Machine Vision and 
Its Application to Structural and Textural Analysis of Remotely 
Sensed Images. ITC Press, Enschede, The Netherlands. 
PAN H. P. (1991b): Formal representation of grammars for spatial 
structure production and its implementation in POP11. Techni- 
cal report, Institut für Photogrammetrie, Universität Bonn, April 
1991. 
PAN H. P. (1991c): Data structures for polygon maps and associ- 
ated manipulation. Technical report, Institut für Photogramme- 
trie, Universität Bonn, July 1991. 
PAN H. P. (1991d): Structural, geometrical and physical modelling 
of landuse maps and images for remote sensing: I — Theory. Te- 
chinical report, Institut für Photogrammetrie, Universität Bonn, 
August 1991. 
  
1. All possible hypotheses (thin lines) generated through prolonga- 
tion from growth vertices (cross in circle) on a segmented image 
(simulated by the grammar). 
  
3. Result after applying the second strong rule: growth vertices 8, 
11, 12, 13 are directly resolved, while vertex 7 demands infor- 
mation from a larger context. 
PAN H. P., W.FORSTNER (1992): Stochastic Polygon Map Gram- 
mars: A generic model for understanding landuse maps and images 
in remote sensing. In: Forstner/Ruwiedel(Eds.), Robust Com- 
puter Vision, Wichmann, 1992. 
QUILLIAN M.R., (1968): Semantic memory. In: M.Minsky (ed.), 
Semantic information processing. Cambridge, Mass.: MIT Press, 
pp.227-270. 
RISSANEN J. (1989): Stochastic Complexity in Statistical Inquiry. 
Series in Computer Science, Vol. 15, World Scientific, Singapore, 
1989 
RITTER H., T.KOHONEN, (1989): Self-organizing semantic maps. 
Biological Cybernetics, Vol.61, pp.241-254. 
RoussEEUW P. J., LERoY A. M. (1987): Robust Regression and 
Outlier Detection. Wiley, N. Y., 1987. 
SANDBERG J., Y.BARNARD (1991): Interview on AI and Education: 
Allan Collins and ‘Stellan Ohlsson. AI Communications 4(4), 
pp.137-138. 
ACKNOWLEDGEMENT 
This work is sponsored by the DFG and supervised by Prof.Dr.habil. 
Wolfgang Fórstner without whose support and guidance this work 
would not have been possible. 
  
2. Result after applying the first strong rule: (growth vertices 1, 2, 
3, 4, 5, 6, 9, 10 are resolved. 
  
4. Search the edge hierarchy: 
level equalities = {A=B,C=D,D=E, E=F } 
level unequalities = { C< A, G<C,H<D,I<E} 
(I, H, G) are brother edges, (A, B) are map boundary edges. 
Fig.2  Error-correcting parsing of segmented image (few important stages) 
937