[pdal] About litree filter: too much computation time, and other questions.

[Ulises Ibarra]

Brad:

Thank you very much for your kind answers.

Ulises

El mar, 17 ene 2023 a las 13:42, Bradley Chambers (<brad.chambers at gmail.com>)
escribió:

>
> On Tue, Jan 17, 2023 at 1:36 PM Ulises Ibarra <
> ulisesmartinibarra at gmail.com> wrote:
>
>>
>> 1. Sampling a point inside a cube with a size of about 10 centimeters on
>> each side, would this seem like a good level of subsampling to you?
>>
>
> I have not thoroughly studied this topic and it will be data dependent,
> but I would guess that this would still work quite well. You might just
> have to experiment on a small patch to tune the settings to your liking
> before expanding to the full scene.
>
>
>> 2. I could cut the cloud of points into tiles, but would something
>> serious happen if a tree is half in one tile and half in another?
>>
>
> Correct. In the best case, you'd have two correctly segmented half trees
> in each tile that you'd need to stitch together somehow.
>
>
>> 3. Would each of the files for each tile be processed by batch processing?
>>
>
>  Any batch processing or parallelization of your workflow will be external
> to PDAL, but would be a reasonable approach to the problem.
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/pdal/attachments/20230117/46c2baf0/attachment.htm>