[GRASS5] [bug #1369] (grass) r.los man page doesn't give a hint how to make the 'pattern map'

[Paul Kelly]

On Tue, 22 Oct 2002, Request Tracker wrote:

> this bug's URL: http://intevation.de/rt/webrt?serial_num=1369
> -------------------------------------------------------------------------
>
> Subject: r.los man page doesn't give a hint how to make the 'pattern map'
>
> Platform: GNU/Linux/i386
> grass obtained from: Mirror of Trento site
> grass binary for platform: Compiled from Sources
> GRASS Version: 5.0pre1
>
> r.los man page doesn't give a hint how to make the 'pattern map' it talks about,
> for us non experts.  Please add a hint.  E.g. first use g.region to make a new region,
> then [how to?] fill it with 1's.  Mention r.mapcalc file=1
>
> -------------------------------------------- Managed by Request Tracker

Try using r.cva (cumulative viewshed analysis) instead of r.los
I just found it at
http://www.ucl.ac.uk/~tcrnmar/GIS/r.cva.html
There is a lot of information and there have been research papers
published about it as well.
It is based on r.los and has so much more functionality and works better.
The instructions that come with it for compiling it are very good; you
just need to change one line in the cmd/Gmakefile from
LIBES = $(GISLIB) $(SEGMENTLIB) $(DATETIMELIB)
to
LIBES = $(DEPGISLIB) $(DEPSEGMENTLIB) $(DEPDATETIMELIB)
and delete or comment the line
char *G_malloc();
from cmd/make_point3.c
I haven't tested it much but using a sites file for the viewpoint didn't
seem to work; I needed to use a raster map with the viewpoint cell having
a value of 1 and all other cells 0. This gave excellent results: a binary
map containing 1 in all the cells visible from the viewpoint and 0
elsewhere. But there are many other possibilities to the program's usage.
The program doesn't use the GRASS5 concept of nulls so you may need to use
r.null to set null values to 0.
Also it doesn't support floating point elevation maps but you can just
e.g. multiply your elevation map by 100 to convert it to centimetres if it
is in metres.

It seems really useful and would only take a bit of work to update it
properly to GRASS5.0 (although it's perfectly usable as it is, with some
care), but maybe there's a reason why it isn't included in the current
GRASS distribution?




>
> _______________________________________________
> grass5 mailing list
> grass5 at grass.itc.it
> http://grass.itc.it/mailman/listinfo/grass5
>