The purpose of the system MOSES is the automatic recognition of objects in aerial images. In this system, a model based structural image analysis is performed. Speci c models are gained through the analysis of digital maps. The models are stored in seantic networks. Image analysis is implemented as a search. To direct this search, one has to evaluate each state of the analysis process. One part of the computed valuations is the model delity, which is a measure for the goodness of match between the choosen image primitives and the speci c model. We present in this article the procedures used to compute the model delity for line segments and polygons. KURZFASSUNG Das System MOSES dient der automatischen Erkennung von Objekten in Luftbildern. Es f uhrt eine modellbasierte, strukturelle Bildanalyse durch, wobei spezi sche Modelle der zu analysierenden Szene durch die Analyse von digitalen Karten gewonnen werden. Die Modelle werden in semantischen Netzen gespeichert. Der Analysevorgang ist ein Suchvorgang, zu dessen Steuerung Bewertungen des aktuellen Analysezustandes anzugeben sind. Ein Teil dieser Bewertungen ist die Modelltreue, die angibt, wie gut die ausgewahlten Bildprimitiven zu dem vorgegebenen Modell passen. In diesem Artikel stellen wir die Prozeduren vor, mit denen die Modelltreue f ur Strecken und Polygone berechnet wird.
[1]
Takashi Matsuyama,et al.
SIGMA: A Knowledge-Based Aerial Image Understanding System
,
1990
.
[2]
Heinrich Niemann,et al.
Control and explanation in a signal understanding environment
,
1993,
Signal Process..
[3]
Franz Quint,et al.
Map—based semantic modeling for the extraction of objects from aerial images
,
1995
.
[4]
Franz Quint,et al.
Colour aerial image segmentation using a Bayesian homogeneity predicate and map knowledge
,
1996
.
[5]
Uwe Stilla.
Map-aided structural analysis of aerial images
,
1995
.
[6]
Richard Gabler,et al.
A Knowledge-Based System for the Analysis of Aerial Images
,
1987,
IEEE Transactions on Geoscience and Remote Sensing.
[7]
Gérard Giraudon,et al.
Multispecialist System for 3D Scene Analysis
,
1994,
ECAI.
[8]
Nicholas V. Findler.
A HEURISTIC INFORMATION RETRIEVAL SYSTEM BASED ON ASSOCIATIVE NETWORKS
,
1979
.
[9]
Glenn Shafer,et al.
A Mathematical Theory of Evidence
,
2020,
A Mathematical Theory of Evidence.
[10]
J. McDermott,et al.
Rule-Based Interpretation of Aerial Imagery
,
1984,
IEEE Transactions on Pattern Analysis and Machine Intelligence.