[pdal] Point cloud generalisation and feature extraction

classic Classic list List threaded Threaded
5 messages Options
Reply | Threaded
Open this post in threaded view
|

[pdal] Point cloud generalisation and feature extraction

Jo Walsh-3
Good evening PDAL list,

I've been at a seminar focusing on machine learning applications for palaeontology, thinking of PDAL and not quite sure why. Various folks here have very high resolution laser or LED scans of very small objects and are interested in doing 3D model classification.

Is there a moral equivalent of GDAL's Polygonize function in PDAL or a Douglas-Peucker like algorithm for point cloud simplification? Is anyone on this list looking at this sort of problem domain? Am I talking out my hat?

Cheers,


Jo
--
  Jo Walsh
  [hidden email]
_______________________________________________
pdal mailing list
[hidden email]
https://lists.osgeo.org/mailman/listinfo/pdal
Reply | Threaded
Open this post in threaded view
|

Re: [pdal] [EXTERNAL] Point cloud generalisation and feature extraction

Newcomb, Doug

On Wed, Sep 12, 2018 at 11:16 AM Jo Walsh <[hidden email]> wrote:
Good evening PDAL list,

I've been at a seminar focusing on machine learning applications for palaeontology, thinking of PDAL and not quite sure why. Various folks here have very high resolution laser or LED scans of very small objects and are interested in doing 3D model classification.

Is there a moral equivalent of GDAL's Polygonize function in PDAL or a Douglas-Peucker like algorithm for point cloud simplification? Is anyone on this list looking at this sort of problem domain? Am I talking out my hat?

Cheers,


Jo
--
  Jo Walsh
  [hidden email]
_______________________________________________
pdal mailing list
[hidden email]
https://lists.osgeo.org/mailman/listinfo/pdal


--
Doug Newcomb - Cartographer
USFWS
551F Pylon Dr
Raleigh, NC
919-856-4520 ext. 14 [hidden email]
---------------------------------------------------------------------------------------------------------

NOTE: This email correspondence and any attachments to and from this sender is subject to the Freedom of Information Act (FOIA) and may be disclosed to third parties.

_______________________________________________
pdal mailing list
[hidden email]
https://lists.osgeo.org/mailman/listinfo/pdal
Reply | Threaded
Open this post in threaded view
|

Re: [pdal] Point cloud generalisation and feature extraction

andrew.bell.ia@gmail.com
In reply to this post by Jo Walsh-3
Hi Jo,

There are quite a few algorithms implemented in PDAL to do data reduction and feature extraction -- it really depends on your needs and the details of your data.  Take a look at the filters listed here: https://pdal.io/stages/filters.html#create

On Wed, Sep 12, 2018 at 11:16 AM Jo Walsh <[hidden email]> wrote:
Good evening PDAL list,

I've been at a seminar focusing on machine learning applications for palaeontology, thinking of PDAL and not quite sure why. Various folks here have very high resolution laser or LED scans of very small objects and are interested in doing 3D model classification.

Is there a moral equivalent of GDAL's Polygonize function in PDAL or a Douglas-Peucker like algorithm for point cloud simplification? Is anyone on this list looking at this sort of problem domain? Am I talking out my hat?

Cheers,


Jo
--
  Jo Walsh
  [hidden email]
_______________________________________________
pdal mailing list
[hidden email]
https://lists.osgeo.org/mailman/listinfo/pdal


--
Andrew Bell
[hidden email]

_______________________________________________
pdal mailing list
[hidden email]
https://lists.osgeo.org/mailman/listinfo/pdal
Reply | Threaded
Open this post in threaded view
|

Re: [pdal] Point cloud generalisation and feature extraction

andrew.bell.ia@gmail.com
In reply to this post by Jo Walsh-3
Should have also mentioned these filters which are designed for data reduction: https://pdal.io/stages/filters.html#cull

On Wed, Sep 12, 2018 at 11:16 AM Jo Walsh <[hidden email]> wrote:
Good evening PDAL list,

I've been at a seminar focusing on machine learning applications for palaeontology, thinking of PDAL and not quite sure why. Various folks here have very high resolution laser or LED scans of very small objects and are interested in doing 3D model classification.

Is there a moral equivalent of GDAL's Polygonize function in PDAL or a Douglas-Peucker like algorithm for point cloud simplification? Is anyone on this list looking at this sort of problem domain? Am I talking out my hat?

Cheers,


Jo
--
  Jo Walsh
  [hidden email]
_______________________________________________
pdal mailing list
[hidden email]
https://lists.osgeo.org/mailman/listinfo/pdal


--
Andrew Bell
[hidden email]

_______________________________________________
pdal mailing list
[hidden email]
https://lists.osgeo.org/mailman/listinfo/pdal
Reply | Threaded
Open this post in threaded view
|

Re: [pdal] Point cloud generalisation and feature extraction

Howard Butler-3
Jo,


I would say there probably isn't, as the problem is very loosely defined. Both of your examples leak too, but not so bad that people can't come up with a 'good enough' solution that most people are satisfied with. 

What do you mean by simplification? Do you mean decimated points (TIN-style, grid-style, others)? Do you mean simplified by having 'similar' points be segmented together?  Do you mean transformed into 3D surfaces with volumes or edges? Andrew gave pointers to all of the possibilities. 

You'll need to be more explicit about what you mean by 'simplified' :)

Howard

PS. Sorry to have missed you in Madison, Wisconsin! Next time you are in the neighborhood, plan to stop by.

On Sep 12, 2018, at 10:40 AM, Andrew Bell <[hidden email]> wrote:

Should have also mentioned these filters which are designed for data reduction: https://pdal.io/stages/filters.html#cull

On Wed, Sep 12, 2018 at 11:16 AM Jo Walsh <[hidden email]> wrote:
Good evening PDAL list,

I've been at a seminar focusing on machine learning applications for palaeontology, thinking of PDAL and not quite sure why. Various folks here have very high resolution laser or LED scans of very small objects and are interested in doing 3D model classification.

Is there a moral equivalent of GDAL's Polygonize function in PDAL or a Douglas-Peucker like algorithm for point cloud simplification? Is anyone on this list looking at this sort of problem domain? Am I talking out my hat?



_______________________________________________
pdal mailing list
[hidden email]
https://lists.osgeo.org/mailman/listinfo/pdal

signature.asc (540 bytes) Download Attachment