Title :
SAR image sampling on DEM generation
Author_Institution :
Canada Centre for Remote Sensing, Ottawa, Ont., Canada
Abstract :
DEM generated from RADARSAT fine mode stereo-images are evaluated as a function of the pixel sampling and also of the terrain relief. Too much undersampling reduces the image content, the geometric and DEM accuracy. Too much oversampling in the image generates a “worms effect”, which decreases the DEM accuracy of about 15%, due to poorer image matching performances. An exact image sampling according to Nyquist law should be then preferred. The DEM accuracy is also linearly correlated with the terrain slopes, and dependent of the terrain aspects, with the best results for the SAR foreslopes and the worse for the backslopes
Keywords :
geophysical techniques; radar imaging; remote sensing by radar; spaceborne radar; synthetic aperture radar; terrain mapping; topography (Earth); DEM generation; Nyquist law; RADARSAT; SAR; backslope; digital elevation model; fine mode stereo-image; foreslope; geophysical measurement technique; image matching; image sampling; land surface topography; oversampling; pixel sampling; radar imaging; radar remote sensing; synthetic aperture radar; terrain mapping; terrain relief; undersampling; worms effect; Data mining; Histograms; Image generation; Image matching; Image resolution; Image sampling; Orbits; Pixel; Solid modeling; Testing;
Conference_Titel :
Geoscience and Remote Sensing Symposium, 2000. Proceedings. IGARSS 2000. IEEE 2000 International
Conference_Location :
Honolulu, HI
Print_ISBN :
0-7803-6359-0
DOI :
10.1109/IGARSS.2000.861706
