[GRASS-user] Closed Areas

GRASS GRASS user
1

Hi,

i transformed my lines into boundaries (v.type) and added than centroids
(v.centroids) to make areas out of them.

GRASS 6.4.0RC5+39438 (BMN):~ > v.type input=FNF output=FNFType
type=line,boundary --overwrite
WARNING: Vector map <FNFType> already exists and will be overwritten
Building topology for vector map <FNFType>...
Registering primitives...
597 primitives registered
1035130 vertices registered
Building areas...
100%
597 areas built
597 isles built
Attaching islands...
100%
Attaching centroids...
100%
Number of nodes: 597
Number of primitives: 597
Number of points: 0
Number of lines: 0
Number of boundaries: 597
Number of centroids: 0
Number of areas: 597
Number of isles: 597
Number of areas without centroid: 597

GRASS 6.4.0RC5+39438 (BMN):~ > v.centroids input=FNFType output=FNFTypeCent
Processing features...
597 new centroids placed in output map
Copying attribute table(s)...
Building topology for vector map <FNFTypeCent>...
Registering primitives...
1194 primitives registered
1035727 vertices registered
Building areas...
100%
597 areas built
597 isles built
Attaching islands...
100%
Attaching centroids...
100%
Number of nodes: 1194
Number of primitives: 1194
Number of points: 0
Number of lines: 0
Number of boundaries: 597
Number of centroids: 597
Number of areas: 597
Number of isles: 597
Number of duplicate centroids: 116
Number of areas without centroid: 116
v.category complete. 597 features modified.

The output says that to 116 areas a centroid can't be added.
Thus i controlled with v.digit, if the boundaries are closed:
http://osgeo-org.1803224.n2.nabble.com/file/n5783822/Screenshot-v.digit_-_FNF%40PERMANENT.png

How you can see, there are just green crosses, that means that every
boundary should be closed? Or not?
But why then there are 116 areas without centroid after runnig v.centroids?

Thanks for any help in advance!

cheers,
pipppoo
--
View this message in context: http://osgeo-org.1803224.n2.nabble.com/Closed-Areas-tp5783822p5783822.html
Sent from the Grass - Users mailing list archive at Nabble.com.

2

pipppoo wrote:

[...]

Number of duplicate centroids: 116
Number of areas without centroid: 116

That looks like there are overlapping areas. For these, a centroid can
not be unambiguously assigned.

v.category complete. 597 features modified.

The output says that to 116 areas a centroid can't be added.
Thus i controlled with v.digit, if the boundaries are closed:
http://osgeo-org.1803224.n2.nabble.com/file/n5783822/Screenshot-v.digit_-_FNF%40PERMANENT.png

How you can see, there are just green crosses, that means that every
boundary should be closed? Or not?
But why then there are 116 areas without centroid after runnig v.centroids?

Because there are also 116 duplicate area centroids. The total number
of centroids is identical to the total number of areas.

You would have to clean the area vector without centroids first with
the tools break,rmdupl, maybe also with rmsa: repeat
tool=break,rmdupl,rmsa until rmsa does no more modifications.

Markus M

Thanks for any help in advance!

cheers,
pipppoo
--
View this message in context: http://osgeo-org.1803224.n2.nabble.com/Closed-Areas-tp5783822p5783822.html
Sent from the Grass - Users mailing list archive at Nabble.com.
_______________________________________________
grass-user mailing list
grass-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user

3

Thanks for you help!

What I really need is one area for every boundary. The reason is that I want
to compute the overlap of the areas, just like done here:
http://grass.osgeo.org/wiki/Vector_Overlapping_Areas
http://grass.osgeo.org/screenshots/vector.php (Overlapping Areas)

So the option with tool:break,rmdupl probably won't work. Is that right?

Has anybody another idea?

This is a closer screenshot of the overlapping areas (randomly colored):
http://osgeo-org.1803224.n2.nabble.com/file/n5783892/Screenshot-Map_Display_1.png

cheers,
pipppoo

--
View this message in context: http://osgeo-org.1803224.n2.nabble.com/Closed-Areas-tp5783822p5783892.html
Sent from the Grass - Users mailing list archive at Nabble.com.

4

I can't solve the problem... :frowning:

When i run (like Markus Metz suggested)

v.clean input=FNFType output=FNFTypeClean tool=break,rmdupl

i get this output (the areas are randomly colorized):
http://osgeo-org.1803224.n2.nabble.com/file/n5784297/Screenshot-Map_Display_1.png
That is, like I mentioned in my previous posting, not what i need.

What i need is is a vector-map, where all closed lines are transformed into
areas. As you can see in the screenshot this happens just for some areas,
but not for all areas (see the white ones):
http://osgeo-org.1803224.n2.nabble.com/file/n5784297/Screenshot-Map_Display_1-1.png
But why? What can I do to transform all closed lines into areas?

ciaociao,
pipppoo
--
View this message in context: http://osgeo-org.1803224.n2.nabble.com/Closed-Areas-tp5783822p5784297.html
Sent from the Grass - Users mailing list archive at Nabble.com.

5

This is a v.digit sceenshot of the vector map:
http://osgeo-org.1803224.n2.nabble.com/file/n5784354/Screenshot.png
What are these "Centroid (duplicate in area)" points? I don't understand
what that means...

Sorry for the many postings, but hopefully they can help to solve my
problem... :slight_smile:
--
View this message in context: http://osgeo-org.1803224.n2.nabble.com/Closed-Areas-tp5783822p5784354.html
Sent from the Grass - Users mailing list archive at Nabble.com.

6

Hi,

2010/11/29 pipppoo <pipppoo@gmail.com>:

This is a v.digit sceenshot of the vector map:
http://osgeo-org.1803224.n2.nabble.com/file/n5784354/Screenshot.png
What are these "Centroid (duplicate in area)" points? I don't understand
what that means...

valid area can have only one centroid inside. More centroids located
within the given area is a topological error.

Martin

--
Martin Landa <landa.martin gmail.com> * http://geo.fsv.cvut.cz/~landa

7

pipppoo wrote:

What I really need is one area for every boundary.

In a topological vector environment such as grass this is not possible
for (fully) overlapping areas.

The reason is that I want
to compute the overlap of the areas, just like done here:
http://grass.osgeo.org/wiki/Vector_Overlapping_Areas
http://grass.osgeo.org/screenshots/vector.php (Overlapping Areas)

Well, you could try to do it just like done there if you first export
the boundary vector (without centroids) to a shapefile and then
reimport the shapefile using the commands in the links above.

Can't think of anything else to get the overlap of the areas.

Markus M

8

Thanks for your help.

Unfortunately the Re-import of the shapefiles doesn't solve the problem.

I will solve the problem in Matlab... :-/
--
View this message in context: http://osgeo-org.1803224.n2.nabble.com/Closed-Areas-tp5783822p5784586.html
Sent from the Grass - Users mailing list archive at Nabble.com.