| Titre | Derivation of Digital Elevation Data Requirements for Operational Satellite Image Geocoding |
| |
| Auteur | Guindon, B |
| Source | Proceedings of the International Symposium on Operationalization of Remote Sensing, International Symposium, Proceedings, Enschede, The Netherlands, April 19-23; 6, 1993 p. 1-13 |
| Année | 1993 |
| Séries alt. | Secteur des sciences de la Terre, Contribution externe 20041115 |
| Séries alt. | RESORS 1093863 |
| Document | livre |
| Lang. | anglais |
| Media | papier |
| Résumé | (disponible en anglais seulement)
Operational geocoding facilities have traditionally generated satellite products which have been projected to a reference ellipsoid. Geometric distortions
associated with terrain elevation variations have been ignored, in large part because of the lack of digital elevation data. With the advent of operational satellite synthetic aperture radars (SARs) and the launch of high resolution, off-nadir
viewing VIR sensors, terrain-related parallax has become a major concern and future operational correction facilities must include large area DEM databases as a routine auxiliary data source. This paper addresses issues related to the design of such
databases. A methodology is presented to quantify and inter-compare the sensitivity of various imaging geometries to elevation displacement. In the design of large area DEM databases it is shown that DEM sampling density can have a significant impact
on geocoding accuracy. From a study of combined global low resolution elevation data (ETOPO-5) and higher resolution DEMs of North America (level 1 DTEDs) it is shown that high resolution spatial sampling requirements can be predicted from ETOPO-5
because of the statistical self-similarity of topography. Example DEM sampling requirement maps are presented for RADARSAT and SPOT imagery. |
| GEOSCAN ID | 217917 |
|
|