Image Information Mining

Automated mapping and information extraction from satellite imagery.

The Image Information Mining conference was the sixth in the series jointly organised by the European Space Agency (ESA) and European Union Satellite Centre (EUSC), and by Joint Research Centre (JRC) since 2009. The conference took place from 3 to 5 November 2009 in Madrid.
This year’s conference continued to focus on automation in support of applications and services for geospatial intelligence, for which Image Information Mining is considered of very high interest and appropriate. This theme proves most successful when cross-cued from other intelligence disciplines, therefore the conference topics were broadened towards more generic Information Mining. In addition, because the future will be marked by an explosion of satellite imaging missions, the conference was expected to bring new stimulating ideas, concepts or methods also for the use of multi-temporal images.

Presentations were focused on theory and applications leading to improve automation in geospatial information extraction and understanding from optical and SAR EO images and heterogeneous sources:

• Automatic image pre-processing (geo-referencing, ortho-rectification, radiometric calibration, etc.)
• Automated feature and information extraction from optical and SAR EO images
• Multi-temporal analysis
• Challenges for metre resolution optical and SAR EO images
• Geospatial Intelligence: synergies across images, maps and geo information
• Models, semantics and spatial syntax for image understanding
• Information mining from heterogeneous sources
• Human-machine communication for spatio-temporal reasoning
• Knowledge discovery and sharing
• Scenarios and constraints in Environment, Security and Intelligence applications
• System architectures for geospatial information processing

Gisat has been selected as one of six European companies and institutions to share its experiences with operational exploitation of satellite imagery. Within the “Bridging the Gap With Operations” section we have presented our results of the joint case study prepared by the conference committee. The study was focused on confrontation and evaluation of automated mapping procedures and processing chains that were applied to analyse QuickBird imagery acquired over South Ossetia in Georgia.

Our results have been highly evaluated by the EUSC and ESA experts. Gisat has again confirmed its position as one of the leading European service providers in the area of land monitoring using remote sensing imagery.