[Therion] Using 3d Scan Data To Create A Map

[kevin dixon]

I do not know.

I have in the past hand cranked a result by plotting LiDAR data 2D
(whole dataset or sub-slices) and manually selecting outer points to
create an outline or cross section but that was circa ten years ago.

An 'automated' approach will need some sort of fractal control -
coastline length problem.

Kevin

On Thu, Jun 20, 2019 at 6:56 PM Martin Sluka via Therion
<therion at speleo.sk> wrote:
>
> Kevin, is any of those softwares able to create vector outline of projection of cloud to a plane?
>
> Martin
>
> > 20. 6. 2019 v 12:43, kevin dixon <geospatial3d at gmail.com>:
> >
> > Those with their own LiDAR instrument will usually have access to
> > their LiDAR manufacturer Software and this often includes automated
> > rendering of man-made surfaces - results quality depend a lot on the
> > LiDAR resolution and accuracy. Natural surfaces such as caves are more
> > difficult for Point Cloud software to deal with automatically.
> >
> > For the rest of us, you may wish to consider the following open source
> > software for Point Clouds:
> > MeshLab - lots of functionality, poor interface, I used this a decade
> > ago when there were few open source options
> > CloudCompare - nice interface, I have not used it for a while
> > ParaView - uses PCL (Point Cloud Library), PCL can be used direct
> > depending on your coding ability
> > PDAL - a library of point cloud tools, need to use something like QGIS
> > for results visualisation
> > Blender - more for 3D videographics
> > 3DTK - 3D Toolkit, not familiar with
> >
> > Other options are available.
> >
> > Kevin Dixon
> >
> > On Thu, Jun 20, 2019 at 9:36 AM Martin Sluka via Therion
> > <therion at speleo.sk> wrote:
> >>
> >> Isn’t it as call for nuclear aircraft carrier to kill a fly?
> >>
> >> Martin
> >>
> >> Odesláno z iPhonu
> >>
> >> 20. 6. 2019 v 9:33, Pavel Herich <herich at speleodd.sk>:
> >>
> >>> This software should do it:
> >>> https://geoslam.com/hub/
> >>> P.
> >>>
> >>>
> >>> Dňa 2019-06-20 09:26 Martin Sluka via Therion napísal(a):
> >>>> I’m curios if there is a software able to create  vector map outline
> >>>> from cloud data automatically.
> >>>> Martin
> >>>> Odesláno z iPhonu
> >>>> 19. 6. 2019 v 23:13, kevin dixon <geospatial3d at gmail.com>:
> >>>>> Nick,
> >>>>> I have both DistoX2 and LiDAR data for parts of the same cave.
> >>>>> All the LiDAR scans were registered to create a single model as normal
> >>>>> for LiDAR.
> >>>>> The individual scans were then decimated in the Point Cloud Software
> >>>>> using a suitable distance to nearest neighbour. I found Survex gets a
> >>>>> bit overwhelmed with millions of LiDAR data so you may have to
> >>>>> experiment with a suitable distance between adjacent LiDAR data. Each
> >>>>> decimated individual scan was then output to LibreOffice Calc and new
> >>>>> ranges, bearings and inclinations calculated to create single station
> >>>>> Survex files with lots of splays, now all orientated the same way.
> >>>>> Dummy traverse observations were created within Survex to link all the
> >>>>> LiDAR scan stations based upon the Point Cloud registration
> >>>>> coordinates and elevations of the scan stations.
> >>>>> I then manually matched the Survex output of the above LiDAR data to
> >>>>> some known DistoX2 observations or in your case, perhaps known show
> >>>>> cave outline. You can do this within Survex by having your non-LiDAR
> >>>>> data as another survey then using a Calibrate Compass figure (may
> >>>>> require up to +/-180degrees) within the LiDAR Survex data to change
> >>>>> the orientation so as to match the known and also setting one of the
> >>>>> LiDAR stations within the Dummy traverse observations to coordinates
> >>>>> and elevation that give a good fit with the known data. Several
> >>>>> iterations are needed with these.
> >>>>> Hope that is self explanatory.
> >>>>> Has worked for me - I have a DistoX2 Survex model with very detailed
> >>>>> LiDAR data in two large chambers - the resulting LOG file looks a bit
> >>>>> weird eg.
> >>>>>   1 8659-node.
> >>>>>   1 19071-node.
> >>>>>   1 21844-node.
> >>>>>   1 22774-node.
> >>>>>   1 23053-node.
> >>>>>   1 24470-node.
> >>>>>   1 25278-node.
> >>>>> As a side note, part of my project includes a Show Cave and I have not
> >>>>> had a problem with insitu metalwork there or on scaffolded sections of
> >>>>> the cave. I have always kept the DistoX2 well away from metalwork and
> >>>>> when the passage is narrow with metalwork, I have used an intermediate
> >>>>> station with DistoX2 observations to it only, never from it.
> >>>>> I have considered the use of turned angles to get through a serious
> >>>>> metaliferous section - the Leica DST360 could do this for you or a
> >>>>> small theodolite - use magnetic obs at each end to orientate it all.
> >>>>> Kevin Dixon
> >>>>>> On Wed, Jun 19, 2019 at 9:02 PM Nick Bairstow <Nick at pff.uk.com> wrote:
> >>>>>> Hi, Something a bit different for the clever people out there.
> >>>>>> I have been playing with some 3d scan data and have attempted to use it to create a Therion map.
> >>>>>> Firstly we put created an outline of the cave passage using point cloud software and exported as dxf.
> >>>>>> I turned the dxf into a background sketch and drew around it and could output a nice pdf of our cave. Also did a manual scaling. So far so good.
> >>>>>> I added a dummy station and gave it entrance coordinates, then exported as a KML model. When I open the output file Google Earth fly’s to the correct spot and shows a pin with the cave name but no cave.
> >>>>>> My question is , does this fail because we have no survey data, centreline etc and it is impossible to do this or am I missing something.
> >>>>>> As an alternative I can export to an esri shp file and run that though Qgis and get what I want but as I am going to attempt to combine the 3 scan data with a several normal disto surveys I would like to see Therion Kml output work if possible to save lots of messing around. (GIS makes my head hurt)
> >>>>>> On a side note using the point cloud software I can slice though the scan data on different planes and pick up lots of detail, rocks and boulders etc. Using the dxf/background image method I should be able to get a very accurate representation of the passage. I can hear you all saying what the hell is he doing that for when he could just survey the cave in the normal way! Well the cave is a show cave and absolutely rammed with metal walkways and cables. Disto tests showed wild inaccuracy’s so it was looking like the only viable option was going to be to use a Total Station device which we don’t have.  We then had a very kind offer to scan the cave, this has proved to be very interesting but drawing a normal map is a bit of a challenge although it seems not impossible.
> >>>>>> Any help would be appreciated.
> >>>>>> Nick
> >>>>>> _______________________________________________
> >>>>>> Therion mailing list
> >>>>>> Therion at speleo.sk
> >>>>>> https://mailman.speleo.sk/listinfo/therion
> >>>>> _______________________________________________
> >>>>> Therion mailing list
> >>>>> Therion at speleo.sk
> >>>>> https://mailman.speleo.sk/listinfo/therion
> >>>> _______________________________________________
> >>>> Therion mailing list
> >>>> Therion at speleo.sk
> >>>> https://mailman.speleo.sk/listinfo/therion
> >>> _______________________________________________
> >>> Therion mailing list
> >>> Therion at speleo.sk
> >>> https://mailman.speleo.sk/listinfo/therion
> >>
> >> _______________________________________________
> >> Therion mailing list
> >> Therion at speleo.sk
> >> https://mailman.speleo.sk/listinfo/therion
> > _______________________________________________
> > Therion mailing list
> > Therion at speleo.sk
> > https://mailman.speleo.sk/listinfo/therion
>
> _______________________________________________
> Therion mailing list
> Therion at speleo.sk
> https://mailman.speleo.sk/listinfo/therion