[GRASS-user] Help for v.lidar.mcc

[Blumentrath, Stefan]

Hi Rafaela,

100mio Points should not be a problem for GRASS in principle.
 
However, in order to be able to answer more specific we would need more info from you, amongst others:
 - output from g.region -up
 - the settings you use in v.lidar.mcc
 - does your point cloud have an attribute table.
 - your system resources (ram and cores)

Generally, I see at least two possible options:

1) You could try v.in.pdal (https://trac.osgeo.org/grass/browser/grass/trunk/vector/v.in.pdal/v.in.pdal.html). Or

2) assuming that memory is the limiting factor, you could subdivide your area of interest into tiles and process each tile (make sure you have enough overlap at the edges).

There might be room for some minor performance improvements in v.lidar.mcc too...

Cheers
Stefan


-----Original Message-----
From: grass-user [mailto:grass-user-bounces at lists.osgeo.org] On Behalf Of Moritz Lennert
Sent: 26. juli 2016 12:17
To: Rafaela Soares <rafaelaniemann at gmail.com>; grass-user at lists.osgeo.org; Newcomb, Doug <Doug_Newcomb at fws.gov>
Subject: Re: [GRASS-user] Help for v.lidar.mcc

On 25/07/16 23:55, Rafaela Soares wrote:
> Hi all,
> I have three point clouds generated from UAV images. I need these 
> filter point clouds, and for this I used the v.lidar.mcc command is 
> separate ground and non-ground.
> So, my new point cloud is more dense. Have more of 100 million points 
> and the v.lidar.mcc seems to be working (this processing to more than 
> one week and not changed, and the computer's performance looks the 
> same in the task manager).
> Can anybody help me? Someone already filtered one point cloud so dense?

https://lists.osgeo.org/pipermail/grass-dev/2016-July/081279.html

I can't help you with an answer, though...

Moritz
_______________________________________________
grass-user mailing list
grass-user at lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user