[GRASS5] [bug #3559] (grass) more resample methods

Request Tracker grass-bugs at intevation.de
Mon Aug 22 14:01:11 EDT 2005


this bug's URL: http://intevation.de/rt/webrt?serial_num=3559
-------------------------------------------------------------------------

Subject: more resample methods

Platform: Mac OSX
grass obtained from: Trento Italy site
grass binary for platform: Compiled from Sources

Either add bilinear and cubic resample commands (ie r.resamp.cubic), or expand r.resample with a 
method option to include these (similar to r.proj).

Up to now I've been able to work around this by using r.proj on a symlink to the location I'm working in, 
so I can used r.proj's bilinear and cubic methods.

But, there is a limitation to this - taking a global projection, such as Robinson, I project it in halves 
from latlong.  This eases memory use, and more important masks out the corners automatically that 
would normally be filled in as a result of the backwards projection method used for rasters.  Then I 
patch the halves together and use the r.proj trick to resample that.  Unfortunately, since I'm resampling 
the whole region, the corners are filled back in.

I generally like to initially project rasters at close to the source resolution, then resample, instead of 
projecting directly to the desired resolution, so I can reuse it for multiple resolutions and save the 
initial projecting and patching time. (in case anyone might suggest I project directly to the desired 
resolution to work around this problem)

-------------------------------------------- Managed by Request Tracker




More information about the grass-dev mailing list