Using gdalwarp (GDAL version 1.6) at the command line.
Input: 8-bit color imagery NITF in geographic projection, pixel sizes of  
4.4956841432e-06 degrees and 4.8692867036e-06 degrees, or very nearly 0.50 
meters per pixel.
Output: intended to be EPSG 3785 at the specified Level 19 pixelsize, which is 
0.2986 m/pixel.

So there is some upsampling involved.  I expected outputsize-pixels to be 
[inputsize-pixels * 0.50/0.2986] in both directions, but it is a good bit 
bigger than that.  Can anybody see why ?  Below is the input & output sizes, 
and the exact command used.

Input size X, Y (pixels) : 45974,  39526
Output size X, Y (pixels) : 83468,  72894

Command used:
gdalwarp -co tiled=yes -r bilinear -t_srs EPSG:3785 -tr 0.2986 0.2986 --config 
GDAL_CACHEMAX 800   input.ntf   output.tif

PS I'd like to query the output file to learn the pixel sizes, but gdalinfo 
doesn't give that for some reason.  Can anyone suggest a means to query that 
info in a BigTIFF ?

.........................................................
Jay Jennings
Senior Software Engineer
www.geoeye.com<http://www.geoeye.com>
703.480.4614
[email protected]
21700 Atlantic Blvd. Dulles, VA 20166
[cid:[email protected]]<http://www.geoeye.com/CorpSite>
Trusted Imagery Experts


<<inline: image001.png>>

_______________________________________________
gdal-dev mailing list
[email protected]
http://lists.osgeo.org/mailman/listinfo/gdal-dev

Reply via email to