[pdal] How to take classifications from one file and put them into another.

Bradley Chambers brad.chambers at gmail.com
Sat Jul 8 21:35:24 PDT 2017


On Mon, Jul 3, 2017 at 5:50 PM Michael Rosen <michael.rosen at gmail.com>
wrote:

I'm doing work on automated classification.  While my input data is all
> LAS, several of the techniques I'm considering currently work only on txt
> files and more significantly some produce labels (classification code) for
> only a subset of the points.  I need to take the labels from classifier and
> put them into the source datasets.  How best to do that?
>

Do I understand correctly that you have one point cloud with
classifications and one without? And that you'd like to apply the
classifications from the first to the second?


> I think a natural way to implement this is to iterate the points in the
> source dataset and for each point, pick the label from the closest point in
> the labelled dataset.  The PDAL's Hausdorf filter does something like this
> and I'm confident I can write a tool that would do this using the KDIndex
> and the Assignment filter.  However, I wonder if there is some way to do
> this as a pipeline using existing filters.
>

I think you can code this up pretty easily using the 3D KDIndex. Hausdorff
would be one example, but there are many others throughout the codebase.
There is no solution that uses existing filters in a pipeline that comes to
mind.

Brad
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/pdal/attachments/20170709/70f4ff92/attachment.html>


More information about the pdal mailing list