Geo-Referencing is something I have zero experience with and I'm needing some help with understanding it.
What is the advantages of Geo-Referencing?
Does the scan have to be registered using target based registration or can it be cloud to cloud and then as long as the targets are within the scan when the surveying data is imported it will pick up on these, this is purely for personal preference as I have never used targets before.
They majority of the work I do is large site jobs that would take a lot of targets which is why I've never used them but a customer's scope has indentified they want geo-referencing as well as co-ordinates aligned to an EPSG, can this be done in SCENE? I thought the idea of a 3D scan was to eliminate the purpose of a traditional survey using Total Stations.
Any advice is muchly appreciated,
Cheers!
Geo-Referencing Point Clouds
-
- I have made <0 posts
- Posts: 7
- Joined: Tue Aug 27, 2019 3:27 pm
- 4
- Full Name: Chris Smith
- Company Details: Primetals Technologies
- Company Position Title: Site Service Engineer
- Country: England
- Linkedin Profile: No
- Has thanked: 1 time
-
- V.I.P Member
- Posts: 352
- Joined: Thu Jan 14, 2016 12:26 pm
- 8
- Full Name: Simeon Herrod
- Company Details: Terra Measurement Limited
- Company Position Title: 3D Data Manager
- Country: UK
- Linkedin Profile: No
- Has thanked: 73 times
- Been thanked: 110 times
Re: Geo-Referencing Point Clouds
95% of our scan jobs need to be to OS coordinates, so we use targets shot in with a TS. Arbitrary is mostly useless to us.
You can also use targets or spheres as just a reference point/plane, without shooting them in.
You can also use targets or spheres as just a reference point/plane, without shooting them in.
- Dave Andrews
- 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: Geo-Referencing Point Clouds
I use survey work a lot on projects that cover a large area. The survey work will keep your scans from drifting or having a banana affect, it will also keep the scans level thought out the project. Another advantage is you can move farther and still maintain accuracy of the survey work, which should be more accurate than the scanner over a long distance.
In scene, you can use the cloud-cloud option as long as the spheres or targets are objects in the scans. Then import the survey data and scene will find the appropriate target for the appropriate survey coordinate. Just make sure the targets are not all on the same line or same elevation, this can confuse scene and your scans might not be properly aligned.
Just an opinion, scanning simply acquires more data than a traditional survey. It does not take the place of survey work and I believe scanning accuracy and efficiency of scanning both field and office are improved by using survey work.
In scene, you can use the cloud-cloud option as long as the spheres or targets are objects in the scans. Then import the survey data and scene will find the appropriate target for the appropriate survey coordinate. Just make sure the targets are not all on the same line or same elevation, this can confuse scene and your scans might not be properly aligned.
Just an opinion, scanning simply acquires more data than a traditional survey. It does not take the place of survey work and I believe scanning accuracy and efficiency of scanning both field and office are improved by using survey work.
"The harder you work, the harder it is to surrender."
Woody Hayes
Woody Hayes
-
- I have made <0 posts
- Posts: 7
- Joined: Tue Aug 27, 2019 3:27 pm
- 4
- Full Name: Chris Smith
- Company Details: Primetals Technologies
- Company Position Title: Site Service Engineer
- Country: England
- Linkedin Profile: No
- Has thanked: 1 time
Re: Geo-Referencing Point Clouds
I never really knew that both could used in conjunction with each other until now, I thought it was more of you do one or the other.Dave Andrews wrote: ↑Wed Nov 18, 2020 1:33 pm I use survey work a lot on projects that cover a large area. The survey work will keep your scans from drifting or having a banana affect, it will also keep the scans level thought out the project. Another advantage is you can move farther and still maintain accuracy of the survey work, which should be more accurate than the scanner over a long distance.
In scene, you can use the cloud-cloud option as long as the spheres or targets are objects in the scans. Then import the survey data and scene will find the appropriate target for the appropriate survey coordinate. Just make sure the targets are not all on the same line or same elevation, this can confuse scene and your scans might not be properly aligned.
Just an opinion, scanning simply acquires more data than a traditional survey. It does not take the place of survey work and I believe scanning accuracy and efficiency of scanning both field and office are improved by using survey work.
How can all this be tied back into an EPSG CRS system?
Cheers.
-
- V.I.P Member
- Posts: 537
- Joined: Mon Jun 16, 2014 1:45 pm
- 9
- Full Name: James Worrell
- Company Details: Bennett and Francis
- Company Position Title: Director
- Country: Australia
- Linkedin Profile: Yes
- Location: Brisbane, Queensland, Australia
- Has thanked: 14 times
- Been thanked: 87 times
- Contact:
Re: Geo-Referencing Point Clouds
Personally I don't like putting ground distance data (ie scanning) on grid distance systems (geodetic) due to the confusion around scaling factors.
As your sites get bigger, the scale factor can become a thing, and with scanners reaching 100's of metres, the scale factor might be an issue. Sure your P40 can scan with a scale factor, but plenty of other instruments can't. It is usually at about this point I see peoples eyes glaze over, which tells me they have no idea about scale factor and to save them from themselves (and my time fixing stuff in the future) they shouldn't be allowed anywhere near geodetic data ;-p
For a house block, sure you can put it on geodetic without much issue (except software support). A site like a university or an airport - the difference between ground and grid can blow out. An airport will likely have surveyors on staff to handle this, but a university or large school? Little chance.
The other issue, geodetic systems move over time - the continents move - so you have different epochs. Then, unless your metadata is good, you might assume data is one system when it is another etc - or worse someone using your data has no clue and doesn't even realise things move. Our MGA94 versus MGA2020 is ~ 1.8m. That is going to get interesting when the "uneducated" start throwing data together in due course. Every single day I see a spec asking for data on "MGA".
Me: "Which MGA do you want the data on?"
Client: "What do you mean which one? There is more than one?"
Me: "FML"
Software also can be problematic - many systems don't handle the large coordinates all that well. Autocad - point cloud osnap won't work for example. Revit - the metric conversion of point clouds to the internal imperial system only uses 4-places of precision - large coordinate point clouds can drop in half a metre out due to the rounding of the translation from the origin. Plenty of software has no idea about EPSG etc - so it is all just numbers anyway - again using Revit, the coorindates have no bearing on anything.
My 2c. Not worth it. We seemed to be building stuff for thousands of years just fine with local coordinate systems. A well documented, thought out, stable control network is going to give you better results.
For "greenfield" sites we use a ground distance, MGA2020-truncated coorindate system - so if necessary we can move data between coordinate systems. Or if all the control is wiped out we can still get back on to it. Or if we want to do GNSS pick up or fly drones or whatever it is all still possible, but the underlying ground distance data is all "local".
You can read my full rant at:
https://www.bennettfrancis.com.au/2020/ ... es-part-1/
https://www.bennettfrancis.com.au/2020/ ... lk-part-2/
(try this one!)
As your sites get bigger, the scale factor can become a thing, and with scanners reaching 100's of metres, the scale factor might be an issue. Sure your P40 can scan with a scale factor, but plenty of other instruments can't. It is usually at about this point I see peoples eyes glaze over, which tells me they have no idea about scale factor and to save them from themselves (and my time fixing stuff in the future) they shouldn't be allowed anywhere near geodetic data ;-p
For a house block, sure you can put it on geodetic without much issue (except software support). A site like a university or an airport - the difference between ground and grid can blow out. An airport will likely have surveyors on staff to handle this, but a university or large school? Little chance.
The other issue, geodetic systems move over time - the continents move - so you have different epochs. Then, unless your metadata is good, you might assume data is one system when it is another etc - or worse someone using your data has no clue and doesn't even realise things move. Our MGA94 versus MGA2020 is ~ 1.8m. That is going to get interesting when the "uneducated" start throwing data together in due course. Every single day I see a spec asking for data on "MGA".
Me: "Which MGA do you want the data on?"
Client: "What do you mean which one? There is more than one?"
Me: "FML"
Software also can be problematic - many systems don't handle the large coordinates all that well. Autocad - point cloud osnap won't work for example. Revit - the metric conversion of point clouds to the internal imperial system only uses 4-places of precision - large coordinate point clouds can drop in half a metre out due to the rounding of the translation from the origin. Plenty of software has no idea about EPSG etc - so it is all just numbers anyway - again using Revit, the coorindates have no bearing on anything.
My 2c. Not worth it. We seemed to be building stuff for thousands of years just fine with local coordinate systems. A well documented, thought out, stable control network is going to give you better results.
For "greenfield" sites we use a ground distance, MGA2020-truncated coorindate system - so if necessary we can move data between coordinate systems. Or if all the control is wiped out we can still get back on to it. Or if we want to do GNSS pick up or fly drones or whatever it is all still possible, but the underlying ground distance data is all "local".
You can read my full rant at:
https://www.bennettfrancis.com.au/2020/ ... es-part-1/
https://www.bennettfrancis.com.au/2020/ ... lk-part-2/
(try this one!)
-
- V.I.P Member
- Posts: 1236
- Joined: Mon Jan 04, 2010 7:51 pm
- 14
- Full Name: Jed Frechette
- Company Details: Lidar Guys
- Company Position Title: CEO and Lidar Supervisor
- Country: USA
- Linkedin Profile: Yes
- Location: Albuquerque, NM
- Has thanked: 62 times
- Been thanked: 219 times
- Contact:
Re: Geo-Referencing Point Clouds
This so much! And let me add "Geoid what's a geoid?" In defense of non-surveyors though I will say they're definitely not the only ones guilty of this. Around here even surveyors who should know better seem to love to assume that everyone works in the same State Plane coordinate system (which variation?) as they do so a csv with coordinates and no additional metadata should be completely self explanatory.jamesworrell wrote: ↑Wed Nov 18, 2020 9:57 pm The other issue, geodetic systems move over time - the continents move - so you have different epochs. Then, unless your metadata is good, you might assume data is one system when it is another etc - or worse someone using your data has no clue and doesn't even realise things move. Our MGA94 versus MGA2020 is ~ 1.8m. That is going to get interesting when the "uneducated" start throwing data together in due course. Every single day I see a spec asking for data on "MGA".
Jed