Download
slide1 n.
Skip this Video
Loading SlideShow in 5 Seconds..
D ata E nrichment for adaptive Gen eralisation Moritz Neun, Robert Weibel, Dirk Burghardt PowerPoint Presentation
Download Presentation
D ata E nrichment for adaptive Gen eralisation Moritz Neun, Robert Weibel, Dirk Burghardt

D ata E nrichment for adaptive Gen eralisation Moritz Neun, Robert Weibel, Dirk Burghardt

167 Views Download Presentation
Download Presentation

D ata E nrichment for adaptive Gen eralisation Moritz Neun, Robert Weibel, Dirk Burghardt

- - - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - - -
Presentation Transcript

  1. Data Enrichment for adaptive Generalisation Moritz Neun, Robert Weibel, Dirk Burghardt Department of Geography, University of Zurich, Switzerland ICA Workshop on Generalisation and Multiple Representation; 20-21 August 2004 - Leicester

  2. outline • project & motivation • data enrichment • horizontal and vertical relations • application using thematic maps, examples • research concept • work plan • first approaches ICA Workshop on Generalisation and Multiple Representation; 20-21 August 2004 - Leicester

  3. project & motivation • DEGEN: Data Enrichment for adaptive Generalisation • the 2 subprojects focus on:1. acquisition of enriched data, representation with MRDBs2. exploitation of enriched data and algorithms • project members:Robert Weibel & Dirk Burghardt (Principle Investigators)Moritz Neun & Stefan Steiniger (PhD Students) • funding: Swiss NSF, 04/2004 – 03/2007 ICA Workshop on Generalisation and Multiple Representation; 20-21 August 2004 - Leicester

  4. project & motivation • application trends:- real-time mapping (web, LBS)- on-demand production • supporting instruments:- data enrichment - MRDB (multiple LODs)- agents • increasing importance of thematic mapping (polygons, POI) ICA Workshop on Generalisation and Multiple Representation; 20-21 August 2004 - Leicester

  5. data enrichment • initial aim: • The purpose of DEGEN is • the data enrichment, the modelling of the enriched data • and the exploitation of this enriched data • focus on: • thematic maps / GIS in a generic way • integration of data enrichment and MRDB for generalisation adaptive generalisation (Galanda & Cecconi) ICA Workshop on Generalisation and Multiple Representation; 20-21 August 2004 - Leicester

  6. data enrichment – terms & definitions • horizontal relationships single level of detail (LOD) only common structural properties (e.g. neighbourhood, pattern, alignment) hierarchic partonomic relations • vertical relationshipslinking of objects and/or object groups on multiple LODs hierarchic partonomic relations ICA Workshop on Generalisation and Multiple Representation; 20-21 August 2004 - Leicester

  7. data enrichment – application domain • thematic polygonal maps: • seamless (tessellation) or • distributed polygons on a common background • important factors: • attributes / semantics • topology • alignments (pattern) • shape • orientation ICA Workshop on Generalisation and Multiple Representation; 20-21 August 2004 - Leicester

  8. geological map detailed: 1:10’000 - 1:50’000 overview: 1:500’000 - 1:1’000’000 data enrichment – map examples ICA Workshop on Generalisation and Multiple Representation; 20-21 August 2004 - Leicester

  9. data enrichment – map examples • geological map • alignment 1:500’000 1:25’000 ICA Workshop on Generalisation and Multiple Representation; 20-21 August 2004 - Leicester

  10. data enrichment – map examples • soil map • broad range from 1:2’000 – 1:1’000’000 1:100’000 1:200’000 ICA Workshop on Generalisation and Multiple Representation; 20-21 August 2004 - Leicester

  11. research concept • work plan • analysis using example maps (patterns, alignments …) • analysis and improvement of existing spatial technologies object model • modelling and managing of the relations in a MRDB data model • automatic identification / generation of horizontal and vertical relations (pattern recognition, object matching …) • exploitation for generalisation e.g. agents, incremental updates, on-the-fly generalisation ICA Workshop on Generalisation and Multiple Representation; 20-21 August 2004 - Leicester

  12. research concept – first approaches • map analysis: • e.g. land use map ICA Workshop on Generalisation and Multiple Representation; 20-21 August 2004 - Leicester

  13. research concept – first approaches • representing islands: • hierarchical and topological relations of two LODs ICA Workshop on Generalisation and Multiple Representation; 20-21 August 2004 - Leicester

  14. map analysis: e.g. geological map research concept – first approaches ICA Workshop on Generalisation and Multiple Representation; 20-21 August 2004 - Leicester

  15. research concept – first approaches • representing alignments: • two parallel alignments of small polygons having the same orientation ICA Workshop on Generalisation and Multiple Representation; 20-21 August 2004 - Leicester

  16. research concept – conclusion • map analysis - current questions: • which relations can be used for data enrichment?(topology, neighbourhood order, orientation …) • how to extract horizontal relations? (pattern recognition, alignments …) • how match objects, object groups?(e.g. pattern, alignments) in different LODs for establishing vertical relations • how represent these relations? MRDB data model neighbourhood order alignment ICA Workshop on Generalisation and Multiple Representation; 20-21 August 2004 - Leicester

  17. thanks • Thanks a lot! • Any questions, suggestions or comments? ICA Workshop on Generalisation and Multiple Representation; 20-21 August 2004 - Leicester