Re: [JPP-Devel] Question on use cases of JTS trianguation API

[Stefan S. <ss...@ge...>]

Hei Martin,

mhm.. need to think about it for a while.

input for my work can bot only be corners of a geometry, but the 
centroid of a geometry as well - if I want to triangulate for instance 
buildings. However, this would only require to derive the centroids as 
an additional step. Import here is the option to attach an "arbitrary" 
object (in my case a jump feature) to a Node (or line) object. I.e. to 
get a link between features (not just geometries) and graph elements.

Breaklines (Collection of LineStrings), mentioned by Sunburned, may be 
interesting too for constrained triangulations. The contour lines of 
course are something else.

In terms of output - what comes to my mind is maybe a bit off-hand but 
of interested in may GIS applications are Voronoi edges and/or Thiessen 
polygons (i.e. the triangulation dual). The latter could be of course 
generated from the Voronoi edges.

It would be also nice to have a function that generates a Minimum 
Spanning Tree out of the triangulation or a proximity graph (I guess the 
latter is a bit different that a triangulation?). I have code on that - 
but not from myself - so I can send it only offlist. If you are 
interested into this too.

Without looking into it - in terms of API would be nice to update the 
triangulation locally. Some more map generalization use cases can be 
found here: 
http://www.ordnancesurvey.co.uk/oswebsite/partnerships/research/publications/docs/2005/034_NICOLASREGNAULD_gen.pdf
N Regnauld 2005: Spatial structures to support automatic generalisation
and here:
http://www.geo.unizh.ch/gis/services/downloads/phd/mneun/phd_moritz_neun_2007.pdf 

... it is in part II - research paper 3: M Neun et al 2008: Web service 
approaches for providing enriched data structures to generalisation 
operators


thanks for your work on this!
I may find more stuff over time ;)

Stefan

Martin Davis wrote:
> The JTS API per se will just be focussed on computing the raw TIN edges 
> and triangles.  Any further information (such as contour lines, 
> slope/aspect etc) are outside the scope of JTS, but can be built fairly 
> easily on top of the functionality the JTS API provides.
>  
> So this won't be replacing the entire JTIN package, but could be used 
> within to the provide the triangulation process.
> 
> My question will probably be clearer when people have had a chance to 
> actually work with the API.
> 
> M
> 
> Sunburned Surveyor wrote:
>> Martin,
>>
>> This is just a suggestion, but Christopher has already done some great
>> work on file formats for TIN information. Maybe JTS could use that?
>>
>> My typical surveying workflow is something like this:
>>
>> Input:
>> - 3D Points (Point Number, Northing, Easting, Elevation, Description)
>> - Breaklines (3D Polyline <AutoCAD> or LineString <JUMP>)
>>
>> Output
>> - TIN
>> - Contour Lines
>>
>> In most cases I'm usually only after the contour lines for display on
>> a topographic map. Sometimes we will use the surface for
>> volume/grading calcs.
>>
>> If you output polygons representing each TIN face, it would be cool if
>> they could have feature attributes representing the slope and aspect.
>> That would allow you to easily to some hillshade rendering and other
>> TIN queries in OpenJUMP.
>>
>> SS
>>
>> On Wed, Apr 29, 2009 at 8:57 AM, Martin Davis <mb...@re...> wrote:
>>   
>>> To those that are interested in the upcoming JTS triangulation API, a
>>> question:
>>>
>>> What type of input and output structures would you find useful?
>>>
>>> Currently I'm developing the following:
>>>
>>> INPUT:
>>> - Geometry (from which the site/vertex coordinates are extracted)
>>> - Collection of Coordinates
>>>
>>> OUTPUT:
>>> - MultiLineString containing triangulation edges
>>> - GeometryCollection of Polygons containing triangles
>>>
>>> You can also work directly with the internal datastructures of the
>>> triangulation (Vertices, QuadEdges, etc), but this requires a higher
>>> level of understanding.
>>>
>>> Is there any other option I haven't though of?
>>>
>>> --
>>> Martin Davis
>>> Senior Technical Architect
>>> Refractions Research, Inc.
>>> (250) 383-3022
>>>
>>>
>>> ------------------------------------------------------------------------------
>>> Register Now & Save for Velocity, the Web Performance & Operations
>>> Conference from O'Reilly Media. Velocity features a full day of
>>> expert-led, hands-on workshops and two days of sessions from industry
>>> leaders in dedicated Performance & Operations tracks. Use code vel09scf
>>> and Save an extra 15% before 5/3. http://p.sf.net/sfu/velocityconf
>>> _______________________________________________
>>> Jump-pilot-devel mailing list
>>> Jum...@li...
>>> https://lists.sourceforge.net/lists/listinfo/jump-pilot-devel
>>>
>>>     
>> ------------------------------------------------------------------------------
>> Register Now & Save for Velocity, the Web Performance & Operations 
>> Conference from O'Reilly Media. Velocity features a full day of 
>> expert-led, hands-on workshops and two days of sessions from industry 
>> leaders in dedicated Performance & Operations tracks. Use code vel09scf 
>> and Save an extra 15% before 5/3. http://p.sf.net/sfu/velocityconf
>> _______________________________________________
>> Jump-pilot-devel mailing list
>> Jum...@li...
>> https://lists.sourceforge.net/lists/listinfo/jump-pilot-devel
>>
>>   
>