|
|
| GeoCommunity Mailing List |
| |
| Mailing List Archives |
| Subject: | Re: [gislist] Align/Match layers from same area |
| Date: |
04/09/2004 09:25:01 PM |
| From: |
Mach Chnife |
|
|
Hey Paul,
I agree with you on that point.
Nevertheless, by saying "most" of the polygons are
very similar, what I really meant to say is: The
majority of polygons are similar. But quite a lot of
other polygons have been updated and are not similar
at all. But that's OK since the algorithm of the
aligning software wouldn't touch those since they're
outside the <0.25m limit. The overall alignment is
somewhat offset probably because of several
manipulations/re-projection of the data.
So what I really want to do is align everything in the
0.25m window and leave alone the rest.
The only plausible automated solution I've found 'til
now is that JCS!! And the best of all: It's totally
free!! I just can't wait that the developers fix the
few bugs I found while executing the "Coverage
alignment" function. I also wish I would know more
about Java so I could give them a hand on this (Since
it is Open Source)!! :-D
Thanks Paul!
mACH nIfE
--- Paul Ramsey <pramsey@refractions.net> wrote:
> Your problem is entirely automatable, the JCS plugin
> for road matching
> does quite similar things. Here is a question though
> -- if the polygons
> are so similar, why bother to do matching at all?
> Choose a reference
> layer (the one you like the spatial qualities of)
> then turn the other
> layer into centroid points and do simple
> point-in-polygon processing to
> transfer the attributes to the reference layer.
>
> The trouble with conflation is that the best
> processes use knowledge of
> the input data in order to make the best matches,
> and each matching
> problem has its own set of inputs, and own desired
> decision process for
> doing matches. Knowing the playing field allows one
> to write really
> good code to automate a maximum quantity of the
> problem. But a priori
> knowledge of the problem mitigates against being
> able to provide COTS
> solutions to conflation problems. COTS solutions end
> up being "better
> than nothing" but a long shot worse than the best
> automated process
> that is possible.
>
> Paul
>
> On Friday, April 9, 2004, at 11:10 AM, Mach Chnife
> wrote:
>
> > Hi Group,
> >
> > I need to align/match two parcels layers of the
> same
> > area but from different sources. Is there a
> sofware or
> > a script/plug-in/add-on to do this?
> >
> > Here is some more details:
> >
> > Most of the polygons on the two layers are the
> same.
> > They just have a slight irregular offset between
> them
> > (<0.25 m). So It would be nice if the new layer
> (A)
> > could be snapped to the more accurate base layer
> (B)
> > we already have in our system.
> >
> > The problem is, intersections of A must snap to
> > corresponding intersections of B: segments of A
> must
> > snap to corresponding segments of B. Sometimes,
> for
> > corresponding segments, A has two vertices but B
> has
> > three (example). Than a vertex must be added to A
> for
> > a proper alignment.
> >
> > Is there any way to do this almost automatically?
> >
>
> Paul Ramsey
> Refractions Research
> Email: pramsey@refractions.net
> Phone: (250) 885-0632
>
__________________________________
Do you Yahoo!?
Yahoo! Mail - More reliable, more storage, less spam
http://mail.yahoo.com
_______________________________________________
gislist mailing list
gislist@lists.geocomm.com
http://lists.geocomm.com/mailman/listinfo/gislist
_________________________________
This list is brought to you by
The GeoCommunity
http://www.geocomm.com/
Get Access to the latest GIS & Geospatial Industry RFPs and bids
http://www.geobids.com
|
|
Sponsored by:
For information
regarding
advertising rates
Click Here!
|
