|
Promut-2 |
| Image processing workflow for GRB mutation and anomaly detection |
| Contact |
Jan Biesemans |
Start (End) Date |
01/09/2009 (01/09/2011) |
| Consortium |
3 partners |
Project Coordination |
AGIV |
| Website |
- |
Related Projects |
SUPERRES |
| Keywords |
photogrammetry, change detection, mobile mapping |
Project Objectives
The N7793 (November 2006 – December 2007), PROMUT-1 (June 2008 – December 2008) and PROMUT-2 (September 2009 – September 2011) projects, are the initiator projects in preparation of the long-term cooperation AGIV-VITO towards a Flemish Spatial Data Infrastructure (SDI).
Methodology
The GRB is the large scale geographic reference database of Flanders containing a set of GIS layers for roads, road elements, buildings, railways, waterways, and parcels. For the continuous update of the GRB several data input channels exist. However, it is expected that photogrammetry will have a very prominent role in the GRB maintenance.
The huge data volumes involved, require a specific long-term vision for a successful integration of photogrammetric data in the GRB maintenance process. Long-term archiving strategies and data preservation, sensor geometric and spectral calibration, compression, image processing algorithms and algorithmic software component maintenance are unavoidable issues when dealing with an image processing workflow for photogrammetric and environmental product generation. This long-term vision has recently (April 2008) been consolidated in a “memorandum of understanding” between AGIV and VITO. As part of these long-term objectives, the short-term IncGEO projects “SUPERRES” (N7793), “PROMUT-1” and “PROMUT-2” aim at creating a prototype workflow for automated GRB change/anomaly detection.
The overall system is subdivided in a number sub-workflows:
(a) Level0 to Level1 calibration support tools: support tools as part of the calibration verification activities for checking the interior and exterior orientation parameters and the radiometric calibration;
(b) Level0 to Level1 workflow: the generation of generic, self-descriptive and standardised Level1 HDF5 files from the raw incoming image data and image metadata in the framework of long-term archiving strategies;
(c) 3D workflow: the automated image-based generation of area-wide digital surface models (DSM) at a user specified spatial resolution;
(d) Level1 to Level2 workflow: the generation of orthorectified and atmospheric corrected single images;
(e) Level2 to Level3 workflow: the generation of mosaic products;
(f) Level2/3 to Level4 workflow: the generation of change detection products compatible with the AGIV GML upload service; and (g) a set of prototype Web Services for the distribution of mobile mapping data to DGI Flanders.
(Expected) Results
A validated and AGIV-reviewed processing system ready for operational use.
VITO Contribution
System development, integration and testing: algorithmic components, middleware, hardware system.
Partners
- AGIV (Belgium)
- VISION++ (Belgium)
- University of Ghent (Belgium)
Illustration
GRB change detection: blue polygons are already mapped buildings, green polygons are the result of an automatic building detection process (combined K-means, Quadratic Discriminant Analysis with Mahalanobis distance metrics and post-classification logic)
