[GRASS-user] Re: v.net.visibility with obstacles that are not included as points

[Wolf Bergenheim]

On Fri, Jun 26, 2009 at 16:03, Moritz Lennert
<mlennert at club.worldonline.be> wrote:
>
> Take the following use case: I want to verify which antennas (points) on a certain site actually 'see' each other and for which visibility is blocked by buildings (areas).
>
> So, I would like to take my point map and submit it to v.net.visibility, telling it that it should use the area map as obstacles. However, v.net.visibility does not work that way: I have to patch the point and the area map and submit the patched map to v.net.visibility, but then all the nodes of the polygons are also integrated into the visibility analysis, i.e. I get the visibility network not only between my antennas, but also between all polygon corners and between the antennas and the polygon corners.
>
> Am I misunderstanding the use of v.net.visibility, or is this something that could be envisaged (i.e. should I post an enhancement ticket in trac) ?

Hi Moritz,

Sorry for the delay, but I've been a bit busy...

The way v.net.visibility works is that it will connect all points that
can be seen to each other. So if you have building polygons, then the
corners of the building that you can see from one antenna are going to
be connected. In your case I'd try it like this.

Run v.net.visibility with only the antennas, then run v.overlay with
the not operator to deselect the visibility lines that are obstructed
by buildings, then the result should be visibility lines between
antennas that can see each other.

Please report how it works.

--Wolf

--

<:3 )---- Wolf Bergenheim ----( 8:>