[GRASS-user] Aggregate a classification map according to a vector
markus.metz.giswork at googlemail.com
Mon Jan 18 07:52:35 EST 2010
Pedro Roma wrote:
> Greetings all
> As far as I can see, r.statistics
> uses as cover and as basis a RASTER map. But, "All cells in the base
> layer are considered one object for the analysis". This is not what I
> want because I want base layer to be a polygon not a cell. Or Am I
> reading this wrong?
You're right, but you can easily convert vector to raster. In this case,
first set the region to match your cover map, then something like
v.what.rast also converts the vector to a raster first, then runs
statistics. v.what.rast would be the best solution unless you want the
most frequent category.
> On Sat, Jan 16, 2010 at 9:59 AM, Markus Metz
> <markus.metz.giswork at googlemail.com
> <mailto:markus.metz.giswork at googlemail.com>> wrote:
> Markus Neteler wrote:
> Pedro Roma wrote:
> Greetings all
> A few emails that people sent, regarding classigication
> operations GRASS
> were extremely helpful :)
> But now, I have one question:
> I need to do some aggregation of the classification map
> according to regions
> specified in a vector file. What I mean is, for each
> county I need to have
> an aggregation of all classified pixels in order to have
> only one value that
> represents that county.
> E.g. I have a land cover classification for my region and
> now I need to
> aggregate the classified map according to my counties in
> order to have a
> vector map in which each county is represented by its more
> Is this possible in GRASS? Was I clear :D ?
> Could this be of interest?
> Addons: r.univar.zonal
> Or r.statistics with method=mode, gives you the most frequent
> value in cover map for each value in base map. AFAIK,
> r.univar[.zonal] doesn't have modal as output.
> Markus M
More information about the grass-user