Logo GISdevelopment.net

GISdevelopment > Proceedings > ACRS > 2000


1989 | 1990 | 1991 | 1992 | 1994 | 1995 | 1996 | 1997 | 1998 | 1999 | 2000 | 2002
Sessions

Agriculture & Soil

Water Resources

Coastal Zone Monitoring

Digital Photogrammetry

Environment

Forest Resources

GIS & Data Integration

Hazard Mitigation

Image Processing

Educational & Profession

Global Change

Landuse

Mapping from Space & GPS

SAR/InSAR

Oceanography

Hyperspectral & Data Acquisition System

AirSAR/MASTER

Poster Sessions
  • Session 1
  • Session 2
  • Session 3



  • ACRS 2000


    Digital Photogrammetry
    Multi-Resolution approach to Radargrammetric Digital Elevation Models Generation

    An example of the correlation at a range of height in the generation of 25m/pixel resolution DEM is shown on figure 12. The locations of the block window centres in the image layers are listed on table 2. It can be seen that a 5m change in height causes a relative shift of 0.22 pixel in the image pair, i.e. 5.5m in distance. As the resolution of the RADARSAT standard image used is 25m, it may suggest that the accuracy of the generated DEM, if the matching accuracy is 1 pixel, will be about 23m.


    Figure 12. Correlation of weighted block windows changes with a range of height

    Table 2. Image blocks and center pixel coordinates for matching.


    The accuracy of the DEM generated from RADARSAT images was evaluated by comparing it with a DEM (fig. 13) generated from Stereo SPOT imagery over the same area. The disparity of the two DEMs is shown in figure 14. The histogram of the disparity is shown in figure 15. The mean height disparity between these two is less than 10 meters, while the standard deviation of the disparity is 26 meters.

    Conclusion
    The imaging model based on physical imaging parameters of the SAR is described. The parameters of our model can be refined with 2 GCPs or more. A weighted block window has been implemented to improve the digital correlation of the matching patches and a hierarchical multi-resolution approach is implemented to increase the robustness of the DEM generation process. The high disparity between the DEM generated from RADARSAT and that from SPOT has a standard deviation of 26 meters. More work will be done to reduce this disparity.

    References
    • Curlander, John C. and Mcdonough, Robert N., "Synthetic Aperture Radar, Systems & Signal Processing", p370-377, 1991.
    • J.S. Greenfeld, "An operator-based matching system", Photogrammetric Eng. Remote Sensing, 57(8), 1049-1055, August 1991.
    • Thierry TOUTIN, "SAR image sampling on DEM generation", IEEE 2000 International Geoscience and Remote Sensing Symposium, volume III, 794-796, July 2000.
    • Willian H. Press, etc., "Numerical Recipes in C", p554.


    Figure 13. DEM generated from SPOT stereopair with IMAGINE Orthomax software.


    Figure 14. The height disparity map between the DEMs generated from RADARSAT stereo pair and SPOT stereo pair.


    Figure 15. Histogram of the disparity map.


    Page 3 of 3
    | Previous |

    Applications | Technology | Policy | History | News | Tenders | Events | Interviews | Career | Companies | Country Pages | Books | Publications | Education | Glossary | Tutorials | Downloads | Site Map | Subscribe | GIS@development Magazine | Updates | Guest Book