In this paper we focus on the application of transferable, object-based image analysis algorithms for dwelling extraction
in a camp for internally displaced people (IDP) in Darfur, Sudan along with innovative means for scientific visualisation
of the results. Three very high spatial resolution satellite images (QuickBird: 2002, 2004, 2008) were used for: (1)
extracting different types of dwellings and (2) calculating and visualizing added-value products such as dwelling density
and camp structure. The results were visualized on virtual globes (Google Earth and ArcGIS Explorer) revealing the
analysis results (analytical 3D views,) transformed into the third dimension (z-value). Data formats depend on virtual
globe software including KML/KMZ (keyhole mark-up language) and ESRI 3D shapefiles streamed as ArcGIS Server-based
globe service. In addition, means for improving overall performance of automated dwelling structures using grid
computing techniques are discussed using examples from a similar study.
KEYWORDS: Data modeling, Image segmentation, Satellites, Image analysis, Process modeling, Image processing, Spatial resolution, Sensors, Knowledge management, Geographic information systems
A rapidly increasing demand for accurate and updated geo-spatial information requires advanced techniques for
extracting and providing relevant information. The presented work was carried out in a 3654 km2 sized area in the region
of Stuttgart/Germany, characterized by high dynamic growth and steady economic development. The project Biotope
Information and Management System (BIMS) provides and monitors aggregated spatial units relevant for regional
planning tasks. We discuss experiences from the first phase of the project, in which we developed an adaptive per-parcel
approach for delineating elementary units using SPOT-5 MS data (5 m GSD). The target geometry was pre-defined by
digital cadastre data from 2005, but not all existing boundaries were retained: some were dismissed, others introduced.
We followed a threefold strategy: (1) a parcel with internal homogeneity remains the same; (2) neighboring parcels with
similar spectral behavior are merged; (3) a single, heterogeneous parcel is split and new boundaries are generated. By
this, the initial number of units dropped to one fourth. The majority of the units were merged due to trans-boundary
homogeneity, one fourth was subdivided. Assets of this approach are its cost-efficiency, the high matching degree of the
produced geometry and the transferability to similar cases because of the standardized character of the data sets involved.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.