CGAL users discussion list

[Xi Cheng <>]

Hi, Sebastien

Thank you very much for your quickly rely.

As you mentioned, the criteria is the most important issue.

Actually,  i have some joint points which the object-lines should crosss them 

and i also made a search algorithm to get it but it would take a bit of times for computation.

so, i am wondering if the CGAL would have the function  which could generate it directly.

But now i realized that the voronoi lines ares separated from each other and it generated

from the Delaunary triangles( i'm not sure for this). so, perhaps i should keep the current method.

Best,

xi

2013/4/11 Sebastien Loriot (GeometryFactory) <>

I think your problem is not easy to solve as long as you don't choose
a criteria for the subgraph to be kept (otherwise the solution is not
unique).
For example, a criteria could be to extract the longest subgraph with
only node with
valence 2.

I suggest to have a  look to classical graph algorithms (like those
implemented in BGL)
to see if one fits your criteria.

Sebastien.


On 04/09/2013 06:49 PM, Xi Cheng wrote:
> Dear all CGAL members ,
>
> I'm using CGAL Segment_Delaunay_Graph for getting the sketch lines and
> central point of polygons. Here are resuls:
>
> 内嵌图片 1
>
> Then, in my second step, i'd like to simplify those voronoi lines into
> only the main lines ( the line could present the shape of polygon in
> most great measure, it commonly would be the lines has longest length ).
> It would like the red lines below:
> 内嵌图片 2
>
> So, i need to filter out some outsider curves, but i'm not familiar with
> the algorithms of Segment_Delaunay_Graph. Any suggestion and reference
> would be sincerely appreciate .
>
> Best,
> Xi


--
You are currently subscribed to cgal-discuss.
To unsubscribe or access the archives, go to
https://sympa.inria.fr/sympa/info/cgal-discuss