[GRASSLIST:2122] Re: i.smap with MASK

[H B]

> classifying with i.smap I use a MASK file as I don't need urban
areas.
> This is what happens after classification: when I rename the MASK to
> it's original file unclassified areas receive a label which is
totally
> unrealistic, by the way (obviously this depends on the pixels that
> sourround the masked-out areas.). The masks are somehow filled up. Of
> course, afterwards you can "clean up" everything...
>
> How can I proceed instead to receive a map with no data areas?


I guess this means i.smap hasn't been updated to deal with NULLs?
If so you should file it as a bug on the GRASS bug tracker:
http://grass.ibiblio.org/bugtracking/bugreport.html

Solution:
Re-write the map with r.mapcalc when the MASK is in place.

g.copy rast=your_MASK,MASK
g.region rast=your_raster_map
r.mapcalc your_raster_map=your_raster_map
g.remove MASK


good luck,
Hamish


__________________________________
Do you Yahoo!?
Find out what made the Top Yahoo! Searches of 2003
http://search.yahoo.com/top2003