DESCRIPTION
r.distance locates the closest points between "objects" in two raster maps. An
"object" is defined as all the grid cells that have the same category
number, and closest means having the shortest "straight-line" distance.
The cell centers are considered for the distance calculation (two
adjacent grid cells have the distance between their cell centers).
The output is an ascii list, one line per pair of objects, in the following form:
cat1:cat2:distance:east1:north1:east2:north2
- cat1
- Category number from map1
- cat2
- Category number from map2
- distance
- The distance in meters between "cat1" and "cat2"
- east1,north1
- The coordinates of the grid cell "cat1" which is closest to "cat2"
- east2,north2
- The coordinates of the grid cell "cat2" which is closest to "cat1"
Flags
-l
The -l flag outputs the category labels of the matched raster objects at the
beginning of the line, if they exist.
-o
The -o flag reports zero distance if the input rasters are overlapping.
NOTES
The output format lends itself to filtering. For example, to "see" lines
connecting each of the category pairs in two maps, filter the output using
awk and then into d.graph:
r.distance map=map1,map2 | \
awk -F: '{print "move",$4,$5,"\ndraw",$6,$7}' | d.graph -m
To create a vector map of all the "map1" coordinates, filter the output into
awk and then into v.in.ascii:
r.distance map=map1,map2 | \
awk -F: '{print $4,$5}' | v.in.ascii format=point output=name separator=space
SEE ALSO
r.buffer,
r.cost,
r.drain,
r.grow,
r.grow.distance,
v.distance
AUTHOR
Michael Shapiro, U.S. Army Construction Engineering Research Laboratory
Last changed: $Date$