I noted that both r.proj/r.proj.seg use floor() in the NN method
(at least in 6.4). Should not be a proper rounding to the nearest integer
more appropriate?
--
Francesco P. Lovergine
I noted that both r.proj/r.proj.seg use floor() in the NN method
(at least in 6.4). Should not be a proper rounding to the nearest integer
more appropriate?
--
Francesco P. Lovergine
Hi Francesco,
I noted that both r.proj/r.proj.seg use floor() in the NN method
(at least in 6.4). Should not be a proper rounding to the nearest integer
more appropriate?
glad you raise the question. I found the same methodology in the java
NN methods and was right these days I was wondering. The docs say it
is due to performance issues. Not sure if it is the same here.
Very interested here,
Regards,
Andrea
--
Francesco P. Lovergine
_______________________________________________
grass-dev mailing list
grass-dev@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-dev
Francesco P. Lovergine wrote:
I noted that both r.proj/r.proj.seg use floor() in the NN method
(at least in 6.4). Should not be a proper rounding to the nearest integer
more appropriate?
No.
If a projected point has grid coordinates (2.99, 3.99), the cell with
indices (2,3) contains that point, and its centre is the closest to
that point.
If the value of cell represents a sample taken at a point, the sample
point is the cell's centre, not its top-left corner.
--
Glynn Clements <glynn@gclements.plus.com>