[GRASSLIST:2733] Re: R.los - scripting analysis, from sites file/data

Paul Kelly paul-grass at stjohnspoint.co.uk
Tue Feb 24 17:42:54 EST 2004


Hello
Not a direct answer to your question but you would do well to look at r.cva:
http://www.ucl.ac.uk/~tcrnmar/GIS/r.cva.html
You don't say what the ultimate goal of your LOS analysis is but r.cva has
some very interesting and powerful modes of operation with specific
application to inter-visibility analyses of archaeological sites. There
are some nice usage examples and screenshots on the above website.

The author, Mark Lake, has written several interesting papers on viewshed
analysis and its application to archaeology, but they are also of general
GIS interest.

In a more direct answer to your question, I seem to recall that the
feature to read the locations from a sites file did not work very well and
it was better to use a raster (e.g. from s.to.rast). But r.los allows to
specify co-ordinates on the command line so you could take the output from
s.out.ascii on your sites list, pipe it into your perl script and work
from there running r.los on one co-ordinate pair at a time. But r.cva is
such a huge improvement on r.los that you should really try and use it if
you can.

Paul

On Tue, 24 Feb 2004, Lars Forseth wrote:

> Hi!
>
> I'm wondering if its possible to script r.los in i the following way:
>
> I have a sites file (archaeological sites, burials (cairns/mounds)) that I
> want to have a viewshed of, and a raster elevation model.  What I want is a
> script that fetches the coordinates from the sites file, makes a viewshed
> analysis, naming the output raster after the name/ID of the site.
>
> Anyone have something like this?
>
> (By the way, could it be done in Perl? - the only programming language that I
> think I know enough of to try to do it in).
>
> regards
> larsf
>
>
> -----------------------------------------------
> Lars Forseth
> Archaeologist
> Nord-Trøndelag Fylkeskommune
> ------------------------------------------------
>
>




More information about the grass-user mailing list