Large Site - C2C + Survey Registration Workflow

Discuss FARO hardware here.
Post Reply
Eppic
I have made 10-20 posts
I have made 10-20 posts
Posts: 16
Joined: Wed Mar 29, 2017 7:50 pm
7
Full Name: David Epps
Company Details: Envision Construction
Company Position Title: Chief Technology Officer
Country: United States
Linkedin Profile: Yes
Location: Atlanta, GA
Been thanked: 1 time

Large Site - C2C + Survey Registration Workflow

Post by Eppic »

I have a large project (50+ acres) that I'm currently working on...lots of buildings, parking decks, and roads/paths between them all. Final goal is an as accurately located composite point cloud as possible...and an eventual conversion to Revit of the site hardscape/paths/topo, decks, and building exteriors. I have 50+ survey control points and pins around the site that I have been setting up adjacent to as part of my workflow and have Northing/Easting/Elevation values for each.

My original methodology has been to divide up the site into logical zones (4-6) that allow me to close loops around buildings, decks, etc., but also that have enough usable overlap between the areas to allow for C2C registration once each zone is compiled. Each zone is in it's own project so that I can work on multiple at once (let one register while I work on clustering another, etc.). Each zone has a varying number of scans...anywhere from 300-600...so the final compiled project will be somewhere between 2000-3000 scans...obviously Scene is going to have a progressively harder time working with it. Some C2C registrations have been taking 24 hours+ even with each zone divided into 25-40 sub-clusters at times.

Here's my actual question...with me having the XYZ data for all of these survey locations, should I be using those points to locate each zone individually or wait until the final project is all combined and register the final cloud to control at that point? I'm worried that C2C will choke on all of those zones with their own individual C2C registration errors already and trying to force the final composite zones to a control network that is spread across all of them will not end nicely.

I also considered using this method to get to a final 'very close' alignment of all of the scans, remove all clusters and essentially run a visual registration but somehow use the control points to lock in 25+ locations around the site while everything else worked to C2C to those locations...I also don't know if Scene can even do that. Kind of like using Ground Control Points with drone photogrammetry to 'pin' spots and have the rest of the data improved by those locations being held in place.

I hope I explained that well enough, but happy to expand if more info is needed...thoughts?
User avatar
Leandre Robitaille
V.I.P Member
V.I.P Member
Posts: 542
Joined: Sat Aug 03, 2019 1:53 am
4
Full Name: Leandre Robitaille
Company Details: Cima+
Company Position Title: Civil Technician - Surveyor
Country: Canada
Has thanked: 61 times
Been thanked: 246 times

Re: Large Site - C2C + Survey Registration Workflow

Post by Leandre Robitaille »

I wouldn't do this c2c for 2 reasons.
It will take long to register, doing c2c of that amount of scans might take forever if you dont run a threadripper computer with inhuman amount of ram. I wouldnt do clusters like you are mentionning as you will see registration error pop up.

But main reason not to do this this way is...scene does not spread error in c2c. It does the c2c, ignores inclinometer values so the project could do a U shape, you might have false positive very hard to fix and have to play lottery or delete certain areas(i.e trees). Get yourself some spheres and do this with target based or use cyclone core for your registration.

Ima just drop this post of mine here; https://www.laserscanningforum.com/foru ... hp?t=18954
Eppic
I have made 10-20 posts
I have made 10-20 posts
Posts: 16
Joined: Wed Mar 29, 2017 7:50 pm
7
Full Name: David Epps
Company Details: Envision Construction
Company Position Title: Chief Technology Officer
Country: United States
Linkedin Profile: Yes
Location: Atlanta, GA
Been thanked: 1 time

Re: Large Site - C2C + Survey Registration Workflow

Post by Eppic »

Leandre Robitaille wrote: Wed Oct 26, 2022 3:44 am I wouldn't do this c2c for 2 reasons.
It will take long to register, doing c2c of that amount of scans might take forever if you dont run a threadripper computer with inhuman amount of ram. I wouldnt do clusters like you are mentionning as you will see registration error pop up.

But main reason not to do this this way is...scene does not spread error in c2c. It does the c2c, ignores inclinometer values so the project could do a U shape, you might have false positive very hard to fix and have to play lottery or delete certain areas(i.e trees). Get yourself some spheres and do this with target based or use cyclone core for your registration.

Ima just drop this post of mine here; https://www.laserscanningforum.com/foru ... hp?t=18954
Thank you for the reply...I've 'played lottery' as you mention so far with a good bit of success just due to the amount of planes and surfaces to use in the C2C registration as I've brought scans into clusters and multiple clusters together...is it perfect? Absolutely not...but the checks I've done have proven to be fairly positive. I'm still concerned about running the C2C against the 4-6 zones and extrapolating the potential error within each amongst each other. Having so many potential survey points to choose from (5-10 per zone) would you say it's possible to survey register each zone individually and then bring them all together that way?
__________________

I also agree 100% with your other thread...I actually stumbled upon a very similar method you mentioned of creating multiple versions of a project, deleting out extraneous points (trees, etc.) and running registration, then bringing back in the deleted points using the revisions folders. Very clever, but a PITA workflow (hope you're listening FARO...this could be very helpful).
User avatar
landmeterbeuckx
V.I.P Member
V.I.P Member
Posts: 1615
Joined: Tue May 01, 2012 5:19 pm
11
Full Name: Lieven Beuckx
Company Details: Studiebureau Beuckx
Company Position Title: Owner
Country: Belgium
Linkedin Profile: Yes
Has thanked: 183 times
Been thanked: 548 times

Re: Large Site - C2C + Survey Registration Workflow

Post by landmeterbeuckx »

If projects are this large i would definately not use scene. Nightmare to work woth with more then a 300 scans.

Take a look at Scantra for registration. I've seen some data of thousands of scans at Intergeo and it is the bomb for registering imho.
LSBbvba
Surveying services - 3D Laserscanning
Tel : +32477753126
www.lsbbvba.be
[email protected]
Eppic
I have made 10-20 posts
I have made 10-20 posts
Posts: 16
Joined: Wed Mar 29, 2017 7:50 pm
7
Full Name: David Epps
Company Details: Envision Construction
Company Position Title: Chief Technology Officer
Country: United States
Linkedin Profile: Yes
Location: Atlanta, GA
Been thanked: 1 time

Re: Large Site - C2C + Survey Registration Workflow

Post by Eppic »

landmeterbeuckx wrote: Wed Oct 26, 2022 6:47 am If projects are this large i would definately not use scene. Nightmare to work woth with more then a 300 scans.

Take a look at Scantra for registration. I've seen some data of thousands of scans at Intergeo and it is the bomb for registering imho.
How does it do with a mix of C2C and survey control points in scenarios like I've described?
User avatar
ProCro
V.I.P Member
V.I.P Member
Posts: 151
Joined: Thu Jun 27, 2019 2:00 pm
4
Full Name: Nino Skupnjak
Company Details: SKIMI64
Company Position Title: procurator
Country: Croatia
Linkedin Profile: Yes
Location: Croatia
Has thanked: 7 times
Been thanked: 30 times
Contact:

Re: Large Site - C2C + Survey Registration Workflow

Post by ProCro »

Scantra lacks c2c, but it has superb plane2plane registration. Also, it's great for buildings or sites where you have any kind of planes, but it's bad on sites with natural features (e.g. archaeology).

On the other side, you have RealWorks, which works great with any kind of data :D
Emanuele
I have made 80-90 posts
I have made 80-90 posts
Posts: 85
Joined: Wed Apr 29, 2015 8:31 am
8
Full Name: Emanuele Ferrari
Company Details: Studio FERRARI
Company Position Title: Titolare
Country: Italia
Linkedin Profile: Yes
Location: Castellamonte (TO)
Has thanked: 9 times
Been thanked: 15 times

Re: Large Site - C2C + Survey Registration Workflow

Post by Emanuele »

Good morning,
I state that I have never managed projects with so many scans, in jobs from 400-500 scans I proceed in this way:
- I create clusters and subclusters
- for some pairs of scans I use common plans (facades, balconies, etc.) to create forced matches
- c2c registration for clusters of 30-40 scans with plan overlap control
- general registration on external targets and common plans
Of each cluster I have several scans with matches in order to check for errors.

In your case you could use the recording of "target" macro clusters using common plans and external references (detected with total station)
VXGrid
V.I.P Member
V.I.P Member
Posts: 544
Joined: Fri Feb 24, 2017 10:47 am
7
Full Name: Martin Graner
Company Details: PointCab GmbH
Company Position Title: Research and Development
Country: Germany
Linkedin Profile: No
Has thanked: 160 times
Been thanked: 175 times
Contact:

Re: Large Site - C2C + Survey Registration Workflow

Post by VXGrid »

Eppic wrote: Wed Oct 26, 2022 9:33 am
landmeterbeuckx wrote: Wed Oct 26, 2022 6:47 am If projects are this large i would definately not use scene. Nightmare to work woth with more then a 300 scans.

Take a look at Scantra for registration. I've seen some data of thousands of scans at Intergeo and it is the bomb for registering imho.
How does it do with a mix of C2C and survey control points in scenarios like I've described?
Pretty well.
It will use them and align everything perfectly. We have a lot of customers (with projects your size) who register everything in Scantra.
But let's ask the Expert.

I'd summon thee, Daniel Wujanz
Eppic
I have made 10-20 posts
I have made 10-20 posts
Posts: 16
Joined: Wed Mar 29, 2017 7:50 pm
7
Full Name: David Epps
Company Details: Envision Construction
Company Position Title: Chief Technology Officer
Country: United States
Linkedin Profile: Yes
Location: Atlanta, GA
Been thanked: 1 time

Re: Large Site - C2C + Survey Registration Workflow

Post by Eppic »

VXGrid wrote: Thu Oct 27, 2022 7:35 am
Eppic wrote: Wed Oct 26, 2022 9:33 am
landmeterbeuckx wrote: Wed Oct 26, 2022 6:47 am If projects are this large i would definately not use scene. Nightmare to work woth with more then a 300 scans.

Take a look at Scantra for registration. I've seen some data of thousands of scans at Intergeo and it is the bomb for registering imho.
How does it do with a mix of C2C and survey control points in scenarios like I've described?
Pretty well.
It will use them and align everything perfectly. We have a lot of customers (with projects your size) who register everything in Scantra.
But let's ask the Expert.

I'd summon thee, Daniel Wujanz
I just got off of a call/high level overview with him re: Scantra actually, haha...great guy...very knowledgeable!

I plan to test Scantra over the coming weeks and see how I can weave it into my workflow...it doesn't sound like it does incorporate any C2C workflows, however there may be a way to use it for the overall registration and then bring it back into Scene for C2C at the end...although it may still break Scene at that point. Time will tell I suppose.
User avatar
sreed
V.I.P Member
V.I.P Member
Posts: 257
Joined: Mon Apr 28, 2014 4:31 pm
9
Full Name: Scott Reed
Company Details: Prologue Systems
Company Position Title: Principal
Country: USA
Skype Name: sabotva
Linkedin Profile: Yes
Location: Richmond, VA USA
Has thanked: 16 times
Been thanked: 15 times

Re: Large Site - C2C + Survey Registration Workflow

Post by sreed »

A HUGE deficiency in Faro Scene is its inability to register C2C while honoring survey control. I've long requested this capability for years and to my knowledge nothing has come of it. For this reason, we use limited C2C in projects with survey control. The only thing Scene can do is a "best fit, rigid transformation" of the C2C cluster to survey control. This may result in cm-level mismatch tensions depending on the quality of the C2C. And when there are high tensions, its time consuming and difficult to determine why and how to fix. Forensic analysis of C2C registration errors is challenging at best.

For your big project, I would acknowledge Scene's dificiency and create numerous small (10-30 scan) C2C clusters that have at least 3 survey references hard named in each. Fortunately you have plenty of control! Then use targeted registration to place them to control using "force by manual name". And with the scale of this project, consider breaking into several or many lsproj. That, afterall, is the power of survey control.

You will have to be clever in how you scan at control points to ensure a single scan isn't required in 2 clusters to ensure both clusters have that single control point. There are workarounds on the desktop if not done in the field, but best practices would have duplicate (or redundant) scans to ensure both C2C clusters contain this necessary survey control.

I'm not sure why one user responded that C2C doesn't use inclinometer. I don't know the source of this complaint as we've seen no evidence of this.
Post Reply

Return to “FARO”