Samuel, I think I know your problem. You might not know your new region coordinates, especialy if say you are going from a lat long to a mercator projection. I wish there was a simple flag in r.proj to account for this, but for a variety of reasons (see archives if you're interested) it can't happen.
Workaround hack:
In location 1, set your region to the raster with
g.region rast=yourmap
Make a vector that conforms to this region with
v.mkgrid map=newvect grid=1,1 position=region
Exit location 1, open location 2 run v.proj for your new region vector
Your region does NOT need to be set to run v.proj, only for r.proj.
Change your region to match the vector with
g.region vect=newvect
Now that your region is set (you will still need to figure out the correct resolution on your own), you can run r.proj.
G'luck
-Ian
On Feb 2, 2005, at 11:02 AM, samuel cavalcante wrote:
Ian,
well, it worked but i have still a question about..
how i set the resolution of the new location to fit the original raster resolution? the way of describe the resolution in the process of creating a new location is quite different of that displayed when i run g.region or r.info (both in the original location)...
thanks in advance
samuel
Ian MacMillan <ian_macmillan@umail.ucsb.edu> wrote:
Samuel,
If you have imported your raster into location 1, then exit that
location. Open up location 2 (your desired projection) and set your
region to match the incoming raster with g.region w= e= s= n=
Following this, use r.proj in location 2 to bring in your raster from
location 1. This is different only if location 1 is an XY location
(without projection info). If so, write back and let us know, or look
at the mail archives.
G'luck
-Ian
On Jan 31, 2005, at 8:02 AM, samuel cavalcante wrote:
> hi ,
> what i have to do to reproject a raster that is in a different
> projection to current location´s projection?
> i have to quit the location an then import this raster in another
> location whith the desired projection?
> thanks
> samuel
>
> Radim Blazek wrote:
> For me it works.
> v.mkgrid map=pok grid=3,2 position=region
> v.extract input=pok output=pok2 type=area list=4
> ...
> Number of boundaries: 8
> Number of centroids : 1
> ...
>
> Radim
>
> Dylan Beaudette wrote:
> > Radim,
> >
> > Thanks for the idea... but doing v.extract type=area leaves me with
> only the
> > centroids for each grid cell...
> >
> > is there anyway to convert a set of line segments into a polygon?
> >
> > thanks!
> >
> > Dylan
> >
> > On Thursday 27 January 2005 12:14 am, Radim Blazek wrote:
> >
> >>Use v.extract type=area
> >>
> >>Radim
> >>
> >>Dylan Beaudette wrote:
> >>
> >>>On Wednesday 26 January 2005 06:09 pm, Hamish wrote:
> >>>
> >>>>>I have used v.mkgrid (GRASS6-CVS) in the past to make vector
> grids,
> >>>>>and all has been well.
> >>>>
> >>>>..
> >>>>
> >>>>>However, this morning I tried to extract a single square from the
> >>>>>output from v.mkgrid, and was only able to get either a single
> >>>>>centroid (almost correct), or the entire boundary (not good).
> Since
> >>>>>the boundaries do not have CATs, there is no way for me to
> extract a
> >>>>>single boundary...
> >>>>
> >>>>..
> >>>>
> >>>>>Any ideas?
> >>>>
> >>>>what about
> >>>>
> >>>> v.category type=boundary option=add
> >>>>
> >>>>to add category numbers to the boundaries?
> >>>
> >>>Thanks! ... this gives the boundaries CATs... but for some reason,
> each
> >>>grid square is made up of about 8 line segments (3 for top and
> bottom
> >>>lines, 1 for each vertical line)...
> >>>
> >>>My hope is that i can make each grid "cell" into a discreet
> polygon, that
> >>>can be used to set the region -- i.e. so that i can accurately
> subsample
> >>>a larger map.
> >>>
> >>>thanks!
> >
> >t;
>
> Yahoo! Acesso Grátis - Internet rápida e grátis. Instale o discador
> do Yahoo! agora.
Yahoo! Acesso Grátis - Internet rápida e grátis. Instale o discador do Yahoo! agora.
