Share Email Print

Proceedings Paper

Knowledge-based analysis of satellite images
Author(s): Klaus-Juergen Schilling; Thomas Voegtle; Peter Muessig
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

This paper shows several working steps for updating the `Digitale Landschaftsmodell 200 (DLM 200)' using satellite images. It is based on a two-step approach: verification and classification. First the existing semantic model (DLM 200) is used for the knowledge based object oriented analysis of the satellite images. At the second stage the information gained from the first step serves to prove and update the DLM 200. Since the DLM 200 is produced by digitizing the map layers of the `Topographische Ubersichtskarte (TUK 200),' typical cartographic aspects have to be considered. Some examples illustrating these effects on a representative class of the DLM 200 are shown. After the determination of these geometric relations between the DLM 200 and the images the `knowledge,' based on the DLM 200, backs up the object based analysis of the satellite images. Image areas which do not fit the DLM 200 are examined at the second stage. The classification has to assign the changes detected in the course of the verification to appropriate classes of the DLM 200. This process uses the parameters of the image analysis as additional information.

Paper Details

Date Published: 17 August 1994
PDF: 5 pages
Proc. SPIE 2357, ISPRS Commission III Symposium: Spatial Information from Digital Photogrammetry and Computer Vision, (17 August 1994); doi: 10.1117/12.182815
Show Author Affiliations
Klaus-Juergen Schilling, Karlsruhe Univ. (Germany)
Thomas Voegtle, Karlsruhe Univ. (Germany)
Peter Muessig, Karlsruhe Univ. (Germany)

Published in SPIE Proceedings Vol. 2357:
ISPRS Commission III Symposium: Spatial Information from Digital Photogrammetry and Computer Vision
Heinrich Ebner; Christian Heipke; Konrad Eder, Editor(s)

© SPIE. Terms of Use
Back to Top