Scanning Using Survey Control

Discuss FARO hardware here.
User avatar
smacl
Global Moderator
Global Moderator
Posts: 1409
Joined: Tue Jan 25, 2011 5:12 pm
13
Full Name: Shane MacLaughlin
Company Details: Atlas Computers Ltd
Company Position Title: Managing Director
Country: Ireland
Linkedin Profile: Yes
Location: Ireland
Has thanked: 627 times
Been thanked: 657 times
Contact:

Re: Scanning Using Survey Control

Post by smacl »

EricPullins wrote: Tue Mar 12, 2019 12:18 pmMaybe I'm missing something, but I'm asking about workflow and what people's process is with using survey control. From what I've read, there really isn't much explanation about what you do to actually work with a survey control. Is is literally as easy as loading in survey data and just letting it do its thing? What kind of work do I need to do to get these survey controls to work their magic?
It depends what form the control takes. If it has been done with a total station, it will be taken in as angles and distances and adjusted to create coordinates with errors at a given level of confidence. These coordinates can be scan targets or scan setup positions. Survey oriented scanning software such as SCC will have full network adjustment as part of the package, for scanning software that doesn't have a survey aspect, you might need a third party package such as STAR*NET. You may also get a number of adjusted coordinates via static GPS, which may also include a mapping scale factor. Once you have scale free, adjusted, control coordinates, these becomes an input to registration to bring all the different scan setups from their own arbitrary grid into the same project grid.
Shane MacLaughlin
Atlas Computers Ltd
www.atlascomputers.ie

SCC Point Cloud module
Raimund58
V.I.P Member
V.I.P Member
Posts: 213
Joined: Tue Jul 03, 2018 8:23 am
5
Full Name: Raimund Klink
Company Details: IHM engineering GmbH
Company Position Title: 3d scanning guy
Country: Germany
Linkedin Profile: No
Has thanked: 193 times
Been thanked: 25 times

Re: Scanning Using Survey Control

Post by Raimund58 »

\o/
User avatar
gordonired
V.I.P Member
V.I.P Member
Posts: 144
Joined: Thu Mar 09, 2017 4:27 pm
7
Full Name: Gordon
Company Details: Roska DBO
Company Position Title: GIS Coordinator
Country: Canada
Linkedin Profile: Yes
Location: Alberta
Has thanked: 25 times
Been thanked: 38 times

Re: Scanning Using Survey Control

Post by gordonired »

For our team, we have found out slowly over time how important survey control is. When we contract it out to surveyors, to make it into useful data in Scene or other software packages, we ask for it to be in an easting/northing/elevation format. This is the same as XYZ, or YXZ depending on the coordinate system. Don't forget that units are important in all of this, and when in hand you can convert it to a .txt file in notepad etc, and bring that in as a point cloud. Some software packages will allow .csv import as well.

You now have several survey points, that can be matched to your points in the point cloud during registration steps.
SAttaya
V.I.P Member
V.I.P Member
Posts: 401
Joined: Mon Mar 23, 2009 4:04 pm
15
Full Name: SAttaya
Company Details: Sev1Tech
Company Position Title: Sr Software Analyst-RemoteSensingEngr
Country: USA
Linkedin Profile: No
Has thanked: 127 times
Been thanked: 17 times

Re: Scanning Using Survey Control

Post by SAttaya »

re: Daniel Wujanz » 12 Mar 2019 05:17

Daniel,

Some questions if you don't mind.
Any that you'd rather not answer for whatever reason, that's fine.

What's the overall dimensions of the area you in your diagram?

If the yellow circles are scan locations, how are they different from the 15 red circles?

some of the red circles are connected by faint red lines, does that mean registration is "poorer"
than the yellow circles connected by the blue lines?

for the 1013 registrations, clusters of clusters with locking of lower level clusters after a successful registration,
or did everything float and the control points just got the standard immense weighting that scene does?

12 green circles, but 24 control points, one high and one low at each green circle,
or something else?

inside (like convention center), outside (around a park or large garden), or a combination of inside and outside?

What method of registration (before applying survey control) did you use:
* target (spheres/quadrants)
* top down follwed by C2C
* on C2C until applied survey control)
* multiple of the above (maybe a guess as to what percent of each type, ie, 20 Targets, 75 c2c and 5 TopDown followed by c2c)?
Which scanner? (120, X series, S series)

any of the clusters registered with targets?

587 scanning stations - Does that mean 587 scans or 587 locations you scanned from?

How did you determine 5.2mm (1/5 inch) absolute 3D accuracy.

Assumed you used inclinometer on all scan?
What about compass and altimeter?
Or did you turn inclinometer off for most or all of the scans?

Thanks In Advance (TIA) for taking the time to answer.
User avatar
Dave Andrews
V.I.P Member
V.I.P Member
Posts: 172
Joined: Thu Oct 18, 2012 8:58 pm
11
Full Name: Dave Andrews
Company Details: R.E. Warner Associates Inc.
Company Position Title: 3d Scanning Coordinator
Country: USA
Linkedin Profile: Yes
Location: Canton, Ohio
Has thanked: 2 times
Been thanked: 22 times
Contact:

Re: Scanning Using Survey Control

Post by Dave Andrews »

Eric,

The workflow is kind of tedious but not difficult. It does not matter if you have targets located before or after you scan. However, depending on the type of targeting you are using will depend on some field workflow changes.

If using Spheres... I will put some kind of physical marker on the ground like a scratch on concrete or a PK nail. During the scanning process you simply need a tripod with a tribrach and sphere setup over the nail. You will need to measure the height of the sphere above the mark so you know where center of the sphere is located for elevation. Then a surveyor can locate the X,Y,Z of the mark. I prefer this method as a total station is much more accurate when measuring to a prism then locating reflector-less.

If using Checker targets... Same ideals basically, only you do not need to measure the height. I use this often, but there is some inherent error using the reflector-less function of the total station.

Once you have the survey data you will need to add the elevation for the center of spheres. For checker targets you will not need to do anything for elevation. For Faro you need to make sure the format is Easting, Northing, Elevation for the XYZ and formatted as a CSV MSDOS file. Then all you need to do is slide the control file into Faro and place it in the cluster you wish to be on control. Update your registration and Faro will auto fit to the control.

If you need any help with this feel free to send me an email.
"The harder you work, the harder it is to surrender."
Woody Hayes
User avatar
Felix_the_Cat
V.I.P Member
V.I.P Member
Posts: 199
Joined: Thu Jun 09, 2016 5:44 pm
7
Full Name: Felix
Company Details: Lockheed Martin
Company Position Title: Senior Member Engineering Staff
Country: USA
Been thanked: 26 times

Re: Scanning Using Survey Control

Post by Felix_the_Cat »

Here are the two ways I use it.

If I just need accurate registration, I hot glue spheres down. I take a whole bunch of points on the spheres with a tracker. I fit the spheres in spatial analyzer and then use the centers to do a more or less manual registration.

If I need to tie in to a coordinate system, I measure some spheres, and get some benchmarks in the same collection. I manually edit the file to create the proper location and orientation. Then after everything is registered you register once more with your control and it applies the transform to everything for you.


Sent from my iPhone using Tapatalk
Image
User avatar
Daniel Wujanz
V.I.P Member
V.I.P Member
Posts: 227
Joined: Fri Aug 24, 2018 11:26 am
5
Full Name: Daniel Wujanz
Company Details: technet GmbH
Company Position Title: 3D Laser Scanning Specialist
Country: Germany
Linkedin Profile: Yes
Location: Berlin
Has thanked: 206 times
Been thanked: 188 times
Contact:

Re: Scanning Using Survey Control

Post by Daniel Wujanz »

Dear Steve,

I have no secrets!
What's the overall dimensions of the area you in your diagram?
Short note: this is a project with more than 6000 stations. The dimensions are approximately 100 x 100 metres. There's also a short scientific article about the project available that I can send you.

If the yellow circles are scan locations, how are they different from the 15 red circles?
All circles are scan locations. The fill colour indicates the 3D-accuracy (more on that later). Green: 1-3 mm, Yellow: 3 - 10 mm, Red: Everything above. The red circles are located in a poorly controlled area due to a unfavourable survey configuration.
some of the red circles are connected by faint red lines, does that mean registration is "poorer"
than the yellow circles connected by the blue lines?
Orange means that the partial redundancies (a measure for the "controlled-ness") are low or 0. In other words: these parts are uncontrolled, so if something funny happens in terms of registration, you're screwed.

for the 1013 registrations, clusters of clusters with locking of lower level clusters after a successful registration,
or did everything float and the control points just got the standard immense weighting that scene does?
This is one cluster! Our customer registered every level (this is the first one) separately and later on threw everything into one big adjustment. In order to avoid tensions in the network we've introduced the stochastics of the tacheometric measurements that would otherwise bias the inner geometry of the net.

12 green circles, but 24 control points, one high and one low at each green circle,
or something else?
I have to ask the customer.
inside (like convention center), outside (around a park or large garden), or a combination of inside and outside?
This is just the inside of a historical building but they also captured the outside. 56,750 m², 13 levels and 1250 rooms.
What method of registration (before applying survey control) did you use:
* target (checkerboards only for stabilization of the block with control points NOT for registration)
* pairwise plane to plane
* block adjustment with planes, control points AND inclinometer data as observations!
* Which scanner? (Z+F Imager 5006, 5010 and Faro Focus 3D X 130 HDR [yet only for the basement which was ~10% of the project])
any of the clusters registered with targets?
No!
587 scanning stations - Does that mean 587 scans or 587 locations you scanned from?
587 locations. Again, this is just one level.
How did you determine 5.2mm (1/5 inch) absolute 3D accuracy.
Variance-Covariance Propagation that stems from a block adjustment. You'll get this number for every station and not just for the stations that are attached to control points.
Assumed you used inclinometer on all scan?

Yes, even though pseudo-axis observations (levelling of the scanner by means of a bubble level) would also stabilize the lot.
What about compass and altimeter?

None of that! This information is way to imprecise. Yet, this information could be used to a certain degree when you initialize the network.
Or did you turn inclinometer off for most or all of the scans?

I didn't capture the data, but this customer always scans with the inclinometer on.

I hope I didn't miss anything and that this helps. If not let me know!

Cheers

Daniel
EricPullins
I have made 10-20 posts
I have made 10-20 posts
Posts: 12
Joined: Fri Oct 12, 2018 4:50 pm
5
Full Name: Eric Pullins
Company Details: TLC Plumbing and Utility
Company Position Title: TLC Plumbing and Utility
Country: United States

Re: Scanning Using Survey Control

Post by EricPullins »

Since we don't own a Total Station, how do you guys get that kind of work done without having the right tools? Do any of you sub out a surveyor?

If any of you do, how does that workflow change? What things do we need to be aware of while working alongside a surveyor?
User avatar
Matt Young
Honorary Member
Honorary Member
Posts: 3929
Joined: Fri Nov 30, 2007 3:03 pm
16
Full Name: Matt Young
Company Details: Baker Hicks
Company Position Title: CAD-BIM Lead
Country: UK
Linkedin Profile: No
Has thanked: 45 times
Been thanked: 40 times

Re: Scanning Using Survey Control

Post by Matt Young »

The ideal is to have at least three good, controlled, targets, in every single scan. But modern auto-register/cloud 2 cloud workflows are drifting away from that. Drift is not what any of us need in a registration... but lets not get into that now ;)

In my mind the most simple workflow for you is to put out some checkerboard targets during your scanning (More targets rather than less). You don't need them in every scan, but you do need to distribute them evenly across the area you are scanning. you also need good overlap between scans.

If you are using a surveyor, then get them to create a control network that includes the checkerboard targets. when naming the targets numbers will do, no need to make it more complicated than it needs to be.

When you register your scans together you can assign the same numbers to each of the targets throughout your scan. The surveyor will supply you coordinates for the targets. If you have reasonable registration results between your overlapping scans. The constellation of targets in your registered scans should match the constellation of the targets supplied by the surveyor.

Be sure to put all auto-registered clusters into a final cluster and apply surveyed control to that. things should not fall apart that way.

If you are intending to be modelling from the point cloud data after registration (Revit for example) you will want to have your data on a local (square to grid) coordinate system. But that local coordinate system may need to be related to a global (not square to grid) system later. If you are working with a surveyor then they will understand how to do this. It's all a matter of having the right coordinates when you finally register the data.

I prefer to register to local first and transform to global in other modelling software if required. Registering to a local square grid requires some detail from the object you are scanning. For example a long straight wall that can be aligned to. If you are lucky then there will be an existing grid that the surveyor can tie in to.

None of this is too difficult - you just have to wade in there on a small project, make a few small mistakes and learn from them.
If you don't see that there is nothing, then you are kidding yourself.
SAttaya
V.I.P Member
V.I.P Member
Posts: 401
Joined: Mon Mar 23, 2009 4:04 pm
15
Full Name: SAttaya
Company Details: Sev1Tech
Company Position Title: Sr Software Analyst-RemoteSensingEngr
Country: USA
Linkedin Profile: No
Has thanked: 127 times
Been thanked: 17 times

Re: Scanning Using Survey Control

Post by SAttaya »

Daniel Wujanz wrote: Wed Mar 13, 2019 9:12 am There's also a short scientific article about the project available that I can send you.
is it https://www.technet-gmbh.com/fileadmin/ ... 7-2018.pdf ???
Post Reply

Return to “FARO”