[GRASS-dev] [GRASS GIS] #2676: r.neighbors on large rasters

GRASS GIS trac at osgeo.org
Thu May 14 09:23:29 PDT 2015


#2676: r.neighbors  on large rasters
--------------------------+---------------------------------------
  Reporter:  dnewcomb     |      Owner:  grass-dev@…
      Type:  enhancement  |     Status:  new
  Priority:  normal       |  Milestone:  7.0.1
 Component:  Default      |    Version:  svn-trunk
Resolution:               |   Keywords:  r.neighbors large rasters
       CPU:  x86-64       |   Platform:  Linux
--------------------------+---------------------------------------

Comment (by dnewcomb):

 Replying to [comment:1 glynn]:
 > Replying to [ticket:2676 dnewcomb]:
 >
 > > In main.c , line 139 I see :
 > > int i , n;
 > >
 > >   I assume that this means that r.neighbors is limited to rasters with
 the number of cells in the 32 bit integer range?
 >
 > If it is, it has nothing to do with above declaration. "i" will never
 exceed the number of outputs, while "n" will never exceed the number of
 cells in the neighbourhood.
 >
 > r.neighbours operates row-by-row, only holding as many rows as are in
 the neighbourhood, and for method=average it isn't maintaining category
 data, so it shouldn't have any problems with rasters with more than 2!^31
 cells. At least, there's no more reason for r.neighbors to have such
 issues than any other raster module.
 >
 > What are the actual dimensions (rows x columns) of the current region?
 (The dimensions of the input map aren't directly relevant).

 Sorry, rookie mistake.

 The region stats are these.

 rows:       1969532
 cols:       1875705
 cells:      3694261020060

 I had used r.external -e to expand the extent of the region to include the
 linked raster.

 Setting the  region to match the raster gives me this size region.

 rows:       440046
 cols:       830958
 cells:      365659744068


 This seems to be working so far.

 Sorry for the noise.

 Kind of curious what is the upper limit for the command, though.

--
Ticket URL: <http://trac.osgeo.org/grass/ticket/2676#comment:2>
GRASS GIS <http://grass.osgeo.org>



More information about the grass-dev mailing list