Skip to content

Lattice nb/wlist help

2 messages · Gary Nobles, David Winsemius

Original
Gary Nobles
# 
hi I am still tryingto do this, I have been working on this for a year but i
have remained stuck...

I have points at regular intervals but within an irregular window

I need to make a nb weights matrix, then nb2wlist 
conceptually a D8 neighbourhood something like this:

0000000
0000000
0011100
0011100
0011100
0000000
0000000

but I also want another matrix like this (D24):

0000000
0111110
0111110
0111110
0111110
0111110
0000000

and so on...

If anyone can help I would be very grateful

Gary

--
View this message in context: http://r.789695.n4.nabble.com/Lattice-nb-wlist-help-tp3253962p3477619.html
Sent from the R help mailing list archive at Nabble.com.
David Winsemius
#
On Apr 27, 2011, at 5:28 AM, Gary Nobles wrote:

            

      
      
      
    

        
Not sure I understand those terms ...

            

      
      
      
    

        
Certainly looks like a matrix

            

      
      
      
    

        
Looked at package raster and the raster objects can be converted to  
matrices but do not accept the same operations with the row() and  
col() functions, so ....

Convert the raster to a matrix. Calling it r1m in this case:

 > as.matrix(r1)
      [,1] [,2] [,3] [,4] [,5] [,6] [,7]
[1,]   NA   NA   NA   NA   NA   NA   NA
[2,]   NA   NA   NA   NA   NA   NA   NA
[3,]   NA   NA    1    1    1   NA   NA
[4,]   NA   NA    1    1    1   NA   NA
[5,]   NA   NA    1    1    1   NA   NA
[6,]   NA   NA   NA   NA   NA   NA   NA
[7,]   NA   NA   NA   NA   NA   NA   NA

#Create a 3 element wide square "sensing" function:

near1s <- function(r, c, mtx) !is.na(mtx[ ( r-1):min((r+1), NROW(mtx)) ,
                                             (c-1):min(c+1,  
NCOL(mtx)) ])

#Run it over your 'matrified' raster:

  for (r in 1:NROW(r1m) ) {
      for( c in 1:NCOL(r1m)){
           newm[r,c] <- sum(near1s(r,c,r1m)) >0}}
  newm
# Not run;
      [,1] [,2] [,3] [,4] [,5] [,6] [,7]
[1,]    0    0    0    0    0    0    0
[2,]    0    1    1    1    1    1    0
[3,]    0    1    1    1    1    1    0
[4,]    0    1    1    1    1    1    0
[5,]    0    1    1    1    1    1    0
[6,]    0    1    1    1    1    1    0
[7,]    0    0    0    0    0    0    0

# Then convert it back to a raster

Somewhere along the line I dropped the use of row and col, so I guess  
this might even work when appropriate modified for the test for sum  
without converting and using the !is.na() function.

Trying that notion:

 > for (r in 1:NROW(r1) ) {
+     for( c in 1:NCOL(r1)){
+          newm[r,c] <- sum(r1[( r-1):min((r+1), NROW(r1)) ,   
(c-1):min(c+1, NCOL(r1)) ]) >0}}
Warning message:
In .doExtract(x, i, drop = drop) : some indices are invalid (NA  
returned)
 > newm
      [,1] [,2] [,3] [,4] [,5] [,6] [,7]
[1,]   NA    0    0    0    0    0    0
[2,]    0    1    1    1    1    1    0
[3,]    0    1    1    1    1    1    0
[4,]    0    1    1    1    1    1    0
[5,]    0    1    1    1    1    1    0
[6,]    0    1    1    1    1    1    0
[7,]    0    0    0    0    0    0    0

The fix seems obvious.