[GRASSLIST:4957] RE: r.le.pixel null() problem

[M.Wegmann]

hi Bill, 

I did the whole process again with different setups:

the r.le.setup is for moving window, enter radius with keyboard, circle ("1.5" up to "8.5" in 1 steps) or square ("2 2" up to "15 15" in 1 steps), default for all other options.

map n_02l is set to null for all values inside a firescar

r.le.pixel map=n_02l sam=m att=b1 (i did it with some div and te1+2 as well, but didn't bother to do it for the different setup tests)

output circles: 1.5 inside firescar is "0", 2.5 up to 8.5 is "null"

outout squares: 2 2 up to 15 15 inside firescar is 0 but the area is getting smaller with increasing radius. the edges are null. 

I checked the pixelvalues inside the firescar in image n_02l, they are null.

hope that helps, thanks Martin



15/11/2002 2:07:15 AM, "William L. Baker" <BakerWL at uwyo.edu> wrote:

>Hi Martin,
>
>I need a little more information to be able to understand
>the bug. I'm the author of r.le programs.
>
>Are you using circles or squares in r.le.setup?  Can you send
>the exact syntax of the r.le.pixel command that produced a
>"0" with 1.5 radius and null with larger radii?
>
>Thanks.
>
>Bill Baker
>
>-----Original Message-----
>From: M.Wegmann [mailto:m.wegmann at web.de]
>Sent: Wednesday, November 13, 2002 8:40 PM
>To: GRASSLIST
>Subject: [GRASSLIST:4944] r.le.pixel null() problem
>
>
>hello, 
>
>it's rather a remark than a question. 
>
>I set all my firescarsvalues on my satellite image to "null()" (r.mapcalc map_out=if(map_in<=x,null(),map_in) )and 
>calcualted mean, sd, diversity etc. with r.le.pixel. When I choose 1.5 pixel radius in the r.le.setup the output inside the 
>firescars which I set to null, were "0". All the calcualtion with radius 2.5, 3.5, 4.5 and 8.5 were "null()" as expected. 
>
>perhaps I am wrong but it looks like a bug ....  or a stupid user ;-)
>
>cheers Martin
>
>
>
>
>
>
>
>
>
>
>
>
>-------------------------------
>
>Martin Wegmann
>Tropical Ecosystem Research Center
>PMB 44 Winnellie / Darwin
>NT 0821
>Australia
>
>.......
>
>priv.
>4 Carpentier Cres.
>Wagaman / Darwin
>NT 0810
>Autralia
>0061 8 8927 1241
>
>
>
>













-------------------------------

Martin Wegmann
Tropical Ecosystem Research Center
PMB 44 Winnellie / Darwin
NT 0821
Australia

.......

priv.
4 Carpentier Cres.
Wagaman / Darwin
NT 0810
Autralia
0061 8 8927 1241