I'm pretty new to GRASS, and would appreciate some help/guidance/suggestions:
I have been smoothing a raster with r.neighbor, as in this example:
r.neighbors input=aland...@permanent output=alandcov_smooth
method=mode size=45 {title=Smoothed}
The raster values are integers representing qualitative categories.
I'd like to modify the smoothing to exclude a particular value.
Specifically, if a cell (is that the right term?) has a particular
value, say 11, then I want to exclude it, as follows:
- if the center of the neighborhood = 11, leave it unchanged (do
not calculate the mode, just return 11)
- if the center is not 11, calculate the mode, but exclude any
points in the neighborhood whose value = 11 (that is, take the mode
only of points not equal to 11)
The real goal is to preserve boundaries between 11 and not 11. This
is a land cover raster; 11 represents water, and I would like the
smoothing to not change water to land or vice versa.
It seems like it might be possible to do it by creating a mask,
smoothing the masked raster, and then putting the 11's back in maybe
using r.mapcalc, but the details are a bit tough for me to put
together, as a GRASS newbie.
Thanks for any suggestions
-Don
--
--------------------------------------
Don MacQueen
Environmental Protection Department
Lawrence Livermore National Laboratory
Livermore, CA, USA
925-423-1062
--------------------------------------
_______________________________________________
grass-user mailing list
[email protected]
http://lists.osgeo.org/mailman/listinfo/grass-user
