Full text: Proceedings, XXth congress (Part 3)

  
  
   
  
   
  
  
    
  
   
   
  
   
   
  
  
  
  
  
  
  
  
  
   
  
  
   
   
  
   
   
  
   
   
   
  
   
  
  
  
  
    
      
  
   
    
    
   
   
    
  
  
    
    
  
  
  
  
   
   
    
ınbul 2004 
ven if the 
ined road 
the region 
nes on the 
ected lines 
ubsequent 
d as road 
Applying 
, we have 
city block 
ing better 
e adaptive 
puting the 
differential 
rrammetric 
in, and H. 
nulti-scale, 
ering and 
ultiple 
RI images 
on Image 
114-118. 
Processing. 
ssachusetts, 
ear feature 
SB-snakes. 
3(8): 985— 
1995. À 
ction From 
17-21 April, 
nn, and H. 
rban Area, 
Aerial and 
Van Gool, 
erlands, pp. 
kes: Active 
iter Vision, 
/erification, 
ote Sensing, 
Idings from 
n of Man- 
A. Balkema 
g by profile 
on of Man- 
. Gruen, O. 
rlag, Berin, 
  
SEMI-AUTOMATIC EXTRACTION OF LINE AND AREA FEATURES 
FROM AERIAL AND SATELLITE IMAGES 
T. Ohlhof* ", E.Gülch^, H. Müller?*, C. Wiedemann *, M. Torre ° 
* ESG Elektroniksystem- und Logistik-GmbH, Einsteinstrae 174, D-81675 Munich, Germany — timm.ohlhof@esg.de 
° University of Applied Sciences Stuttgart, SchellingstraBe 24, D-70174 Stuttgart, Germany — 
eberhard.guelch@hft-stuttgart.de 
* Snowflake Software Ltd, 8a Carlton Crescent, Southampton, SO15 2EZ, United Kingdom — 
hardo.mueller(gsnowflakesoftware.co.uk 
? MVTec Software GmbH, Neherstrafe 1, D-81675 Munich, Germany — wiedemann@mvtec.com 
© Institut Cartografic de Catalunya, Parc de Montjuic, E-08038 Barcelona, margat@icc.es 
Commission III, WG I11/4 
KEY WORDS: Automation, Extraction, GIS, Imagery, Object, High resolution, Vector 
ABSTRACT: 
The aim of the project, which was carried out by the two software companies ESG and Inpho as well as by the Chair for 
Photogrammetry and Remote Sensing of the Technische Universitit Miinchen and the Institut Cartografic de Catalunya (ICC) was to 
develop an operational software system for the semi-automatic extraction of line and area features in 2D and 3D from aerial images 
and high resolution satellite imagery. The complete system was delivered to the Geo-Information Office of the German Federal 
Armed Forces and has been in practical use since May 2003 for the update of VMap Level | vector data and the generation of the 
military basic vector database. The extraction tools for line and area features are integrated into Inpho's software platform inJECT, 
which was originally designed for the measurement of 3D building models in digital imagery. For the import and export of the GIS 
vector data an interface between inJECT and the GIS packages Dynamo and GeoMedia (Intergraph) has been developed based on the 
GML2 format standard. 
The semi-automatic extraction is preferably done in digital orthophotos for the capture of 2D GIS vector data. In addition, the 
software is available for the capture of 3D features using oriented aerial imagery. The algorithms and workflows have been extensively 
tested with IKONOS 2 and IRS satellite imagery as well as orthophotos with 50 cm pixel size. 
1. INTRODUCTION 
The paper deals with the semi-automatic extraction of line and 
area features from aerial images and high resolution satellite 
imagery. In practice, the extraction of topographic objects from 
images for generating and updating GIS databases is carried out 
interactively based on mono or stereo plotting. Numerous 
efforts have been made in the past to automate the acquisition of 
point, line, and area features in aerial imagery (Gülch 2000, 
Baltsavias et al. 2001, Ohlhof et al. 2000). Fully automated 
(autonomous) systems, however, are until now in the research 
stage or can be only used for limited purposes (Baltsavias et al. 
2001). Semi-automatic systems assisted by an operator seem to 
be the best solution for the near future (Gülch et al. 2000, Gülch 
& Müller 2001, Inpho 2004, CyberCity 2004, Ulm 2002). 
The aim of the project, which was carried out by the two 
software companies ESG and Inpho as well as by the Chair for 
Photogrammetry and Remote Sensing of the Technische 
Universität München (TUM) and the Institut Cartografic de 
  
Corresponding author 
Catalunya (ICC), was to develop an operational system for the 
semi-automatic extraction of line and area features in 2D and 3D 
based on an existing software platform. The complete system 
was delivered to the Geo-Information Office (AGeoBw) of the 
German Federal Armed Forces and has been in practical use 
since May 2003 for the update of VMap Level 1 data and the 
generation of the military basic vector database. 
The paper first gives an overview of the software architecture of 
the extraction system and its components. Some implementation 
issues are mentioned. In the next section, the data base and the 
extraction workflows are described. The methods and algorithms 
for the GIS interface and the extraction tools are the subject of 
the main section. Results achieved with high resolution satellite 
imagery and orthophotos are depicted. Finally, conclusions are 
drawn. 
  
	        
Waiting...

Note to user

Dear user,

In response to current developments in the web technology used by the Goobi viewer, the software no longer supports your browser.

Please use one of the following browsers to display this page correctly.

Thank you.