Thank you Michael. Good to know about that capability of the georectify tool.
Yesterday Maciej emailed me a different solution, which also worked. I thought my question was so elementary that users on this list would not want it posted. But he suggested that I do that, for the benefit of other newbies, I suppose. So here is a summa:
- The shapefiles had .proj files associated with them. I used one to create a UTM location in GRASS.
- The .sid raster had no projection info attached to it (so far as I can tell). I used it to create an x,y location.
- The extents of the x.y location were broader, and the cell resolution much higher than the UTM location. I used g.region to increase the extents and resolution of the UTM projection to match the x,y location (and presumably the .sid raster).
- I imported the .sid raster directly into the UTM location. Actually, I only imported the red band (band=1) because that alone was 1.7 GB and ground features resolved most clearly in that particular band.
- The imported vectors and raster line up, so at least I have replicated the registration of the original ArcGIS coverage.
So in this procedure I only created the x,y location to extract information from the .sid file: its extents and resolution. I never pulled data into the x,y location. Thank God, because that raster takes 2 hours to import into any location!
Now struggling with v.overlay of intersecting buffers over polygons…grrrr…
Ciao,
Pietro Calogero
http://www.calogero.us/Kabul/2007/
On Sep 22, 2007, at 9:00 PM, Michael Barton wrote:
Pietro, see below
On 9/20/07 8:56 PM, “Pietro Calogero” <pietro@calogero.us> wrote:
I received a set of data that was developed in ArgGIS, including shapefiles
and a .sid raster file. The shapefiles came with projection information, but
the .sid file did not (and it extends out further than the vector info on the
edges). So I created an x,y location for the .sid file and imported it using
GDAL. It imported fine; the satphoto looks good. I also created a UTM location
for the shapefiles using their proj info, and then extended the region and
increased the resolution based on the extents of the raster (all units are in
meters).
If you KNOW the projection of the .sid file AND it IS projected (just
doesn’t have a header with projection information) you can initially create
a proper location and simply import the file overriding its (lack of)
projection information. This is a check box in the GUI and the -o flag on
the command line. No need to reproject the file.
Now I want to reproject the raster from the x,y location into the UTM
location. Both are in meters, both have the same N,S,W,E extents, both have
the same resolution.
If you have the file in an xy location and want project it into a UTM
location, the simplest way is to use the georectification module in the GUI.
Start GRASS in the proper UTM location. Make sure that you set the region
extents and resolution to match the image extents and native resolution for
best results. Use g.region to do this.
First, display one of your vector maps in the normal way.
Start the georectification module under the file menu.
- select the location/mapset of the file you want to georectify
- create a georectifcation “group” if you have not already defined one.
Your group will contain only 1 file
- select that group
- select a map/image to display for georectification. In your case, it is
your single image.
- start georectification
If you don’t know the correspondence between xy points and desired UTM
equivalents, you could interactively click a point on your image and a
corresponding point on your vector to get that information for a series of
points.
However, in your case, it sounds like you already know that there is a 1:1
correspondence between x/y values and UTM east/north and your map is already
rectified (i.e., doesn’t need any kind of warping). If so, you can just type
the corner coordinates into the xy box (separate by commas as in
xcoord,ycoord) and the same number into the eastnorth box (format as
east,north). Do this for the four corners, pick 1st order rectification and
click the georectify button.
Hope this helps.
Michael
Michael Barton, Professor of Anthropology
Director of Graduate Studies
School of Human Evolution & Social Change
Center for Social Dynamics & Complexity
Arizona State University
phone: 480-965-6213
fax: 480-965-7671
www: http://www.public.asu.edu/~cmbarton