Dear list,
I am (again :)) buffering a cloud of points ("buffering" as in spatial
analysis sense).
The step from MultiPointFeature to (Multi)Polygon rapidly becomes a
performance bottleneck as I have to repeat this operation several times per
second.
I am keeping performance scalable by segmenting the points as my cloud gets
updated so that I only have to process a limited number of points.
The little drawback is that overlapping "artifacts" appear here and there
where different features ("segments") overlap -- (fill color has
transparency). But I don't mind this.
I am wondering if there is some technique that can further bring down the
cost of the buffering, especially with respect to the current map scale.
The density of the points is pretty high so an ad-hoc simplification of the
point cloud might be a good thing.
At a first I do not see remarkable gains whenever I apply one of these JTS
simplifiers (Douglas-Peucker or TPS), but maybe I am misusing them.
Or maybe this is not exactly the use-case their are meant for. ?
I'd just like to dig a little more.. Has anyone experience with JTS
simplifiers?
Any other practical suggestions for my case?
Thank you,
-Piero
------------------------------------------------------------------------------
_______________________________________________
GeoTools-GT2-Users mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/geotools-gt2-users
