Georeferencing (Sorry)

Discuss all FARO related issues here.
User avatar
ddustin
V.I.P Member
V.I.P Member
Posts: 931
Joined: Mon Aug 22, 2011 9:28 pm
Full Name: David Dustin
Company Details: FARO Technologies Inc
Company Position Title: Global Dir of PS Product Management
Country: USA
Linkedin Profile: Yes
Location: Georgia, USA.. Planet Earth

Georeferencing (Sorry)

Post by ddustin » Wed May 30, 2018 2:22 pm

Group,
I have searched through the database here for clearly defined methods for Georeferencing scan data.

Forgive my ignorance, but I am trying to learn this for a specific reason.
My questions are:
1. When we import GPS coordinates should it be in the form of .txt, .cor or .csv (is one of them better than the others and if so, why?)
2. What is the proper format for the imported coordinates? Name,XComponent,YComponent,ZComponent (where do we get the Z component if they are GPS Coords?)
3. Should the GPS Coordinates that are being provided (from military grade devices) be converted to UTM before importation?
4. If the data is converted to UTM before importation, do we import all the components (what do we do with the "zone" information?)?
5. Once we have our scans georeferenced should we move clusters to center of scans?
6. How do I verify my scans are correctly georeferenced?
7. If I want to know the GPS coordinates of a specific location in my scans, how do I do that? (if I create a point and look at it's properties, the GPS values are not there and if we enable GPS for that point, it appears to be a reference value for the entire scan as every other point in the scene has the very same coords)

I'm sure I'll have more questions, but thank you in advance for any advice.

David
Global Director of Public Safety - Product Management

User avatar
Scott.Warren
I have made 30-40 posts
I have made 30-40 posts
Posts: 33
Joined: Tue Jan 30, 2018 3:14 pm
Full Name: Scott Warren
Company Details: WSP USA
Company Position Title: Project Manager - Geomatics Engineer
Country: USA
Linkedin Profile: Yes
Location: Merrimack, NH, USA

Re: Georeferencing (Sorry)

Post by Scott.Warren » Wed May 30, 2018 10:04 pm

I think you need to do some reading up on Coordinate systems and Datums. It would help because your questions are using the GPS a bit too freely.

GPS = Global Positioning System
A GPS unit collects information from satellites and give you a latitude, longitude and and elevation in the World Geodetic System 1984 (WGS84). This Datum, WGS84, is represented most often in Lat and Long. You mentioned UTM, which is a mapping projection which can translates the lat, long, to a linear unit (ft/m). UTM is one of many ways to represent spatial information. There are many datums and map projections that can be used to make maps. Datum and Map project would be useful terms to read into. If your asking what Zones are in UTM, its clear that you need do a little homework.

You should stop thinking about your coordinates as GPS coordinates. Think of them as coordinates in datum X. A GPS unit can transform its co-ords from WSG84 into almost any other datum on the fly, or with some processing. Saying your points are on GPS coordinates could mean many things, saying your co-ords are in, for example, NAD83 Horizontal Datum and NAVD88 Vertical Datums is concrete and clear to your peers.

Having a solid foundation on understanding in Datums/coordinate systems/mapping projections is key to geo-referencing correctly. Knowing what coordinates your final result should be in will make clear what coordinates derived from GPS observations should be used in your registration.

GPS data collection methods can take many forms. "Military grade Device" doesn't help narrow down what you mean. There are special Military GPS signals that I would imagine a civilian unit would not have access to, and there are "military Spec" IPV ratings for some GPS units, but the end result should not be that different from the GPS used by the survey industry (in terms of output). Could you be more specific on this point?

I'll leave the FARO registration process to others, as it is out of my comfort zone. I'll just note that if registration is done correctly, the values of each point within the point cloud should fall within the chosen datum, so every point will be accurately represented in that Datum. Meaning, if a surveyor were to perform a field survey after you scanned, their measurements (X,Y,Z) on a feature should match those of the same feature within the pointcloud (X,Y,Z).

You asked a lot of questions, and I hope I was able to help in some small way.

Graham
V.I.P Member
V.I.P Member
Posts: 147
Joined: Mon Apr 27, 2009 3:38 pm
Full Name: GrahamM
Company Details: AVEVA
Company Position Title: Software Engineer

Re: Georeferencing (Sorry)

Post by Graham » Thu May 31, 2018 11:09 am

David,

As long you document the projection used you will be doing better than most. Personally selecting a coordinate system to be universal is fraught with danger of somebody just taking a contrary position. For example some might say using UTM developed by the military so they could map world so they know how to move troops, bombs etc. So the accuracy of the projection is correct for their needs.

It is difficult as metadata on surveys seems to be the first thing that is lost, if we could embed this data into the scan it would be best but we are were we are.

I recommend speaking with your client to clarify what they think it means. In my experience of Plant design they user wants to work in Plant coordinates, they really do not want big numbers and they want to align to Plant north. Purchasing departments who have no connection with Design dept. specify the data to be geo-referenced and then all hell is let loose. Before anyone says well Design systems should be able to cope, they increasing are but that does not mean customer will pick up that version of the product. As user will not necessarily upgrade there software mid-design, why should they it works changing it only introduces risk.

Oatfedgoat
Global Moderator
Global Moderator
Posts: 1018
Joined: Mon Oct 06, 2008 8:31 pm
Full Name: Matthew McCarter
Company Details: Costain
Company Position Title: BIM Manager
Country: England
Linkedin Profile: Yes

Re: Georeferencing (Sorry)

Post by Oatfedgoat » Thu May 31, 2018 12:29 pm

Graham wrote:
Thu May 31, 2018 11:09 am


It is difficult as metadata on surveys seems to be the first thing that is lost, if we could embed this data into the scan it would be best but we are were we are.
It would be fantastic if software systems started to introduce a field of metadata for accuracy.
I would go as far as wanting it to be an attribute to the file, whether that file is a processed traverse in starnet, a processed survey in N4ce etc, A topo survey in AutoCAD or a full BIM in Revit.
Maybe e57 is the only likely option where it could happen for point clouds?

This would massively help ensure that the accuracy and therefore useability of a survey is never lost as well as help with audit and assurance trails/reports.

User avatar
3DForensics
Global Moderator
Global Moderator
Posts: 1797
Joined: Mon Aug 03, 2009 1:52 am
Full Name: Eugene Liscio
Company Details: AI2-3D Forensics
Company Position Title: Owner
Skype Name: eliscio
Location: Toronto, Canada

Re: Georeferencing (Sorry)

Post by 3DForensics » Thu May 31, 2018 12:55 pm

Great post guys and great information. As a person who is usually interested in local, relative measurements, I am quite interested in what GPS can do for us.

David, your question about locating an object in GPS coordinates is a really good one and I would also like to know the answer to that one.

One thing I have heard which I would like to verify...is elevation in a GPS system usually the one with the greatest error?

Eugene

User avatar
Scott.Warren
I have made 30-40 posts
I have made 30-40 posts
Posts: 33
Joined: Tue Jan 30, 2018 3:14 pm
Full Name: Scott Warren
Company Details: WSP USA
Company Position Title: Project Manager - Geomatics Engineer
Country: USA
Linkedin Profile: Yes
Location: Merrimack, NH, USA

Re: Georeferencing (Sorry)

Post by Scott.Warren » Thu May 31, 2018 1:22 pm

For GPS accuracy, that depends a lot on the type of GPS tech you are using, not as much on make/manufacturer as they are all receiving the same signals from OUTER SPACE. The 2 main types used by surveyors are called RTK, and Static. Static involves using multiple GPS units collecting information at the same time, over a long period (30min - 4 hours). Having redundant measurements allow for corrections to the way the GPS signals behave in the atmosphere, giving you Approx 1cm of error in X,Y,Z. RTK (real time kinematic) is a method of using 2 GPS units, one stationary (base), and one on the move (rover). These units are connected in 'real time' by radio telemetry (could be cellular, could be radio carrier), which allow for similar coorections to the GPS signals, but in real time. This is the tech you seen on Mapping Drones, as the drone itself is a rover, and is connected to a GPS base station somewhere. RTK is typically accurate to 2-3cm. A GPS from a cellphone, or any other Non-Survey/Mulitary GPS will achieve an accuracy of about 0.70m (70cm)

https://www.gps.gov/systems/gps/performance/accuracy/

As for Elevations with GPS, they do have more error in the Z direction, as the X-Y Plane. This is due to the angle of the GPS satellites, and how they only appear above the user, not to the sides, or from underground. If you imagine the user as the center of a sphere, the GPS signals they receive only come from a cone shape starting at the user, and spreading upward and outward. If they were able to receive signals from underground (magically), the X-Y plane and Z direction would have the same/similar error. This occurs in OUTER SPACE when using GPS, as there are situations where the GPS signals can come from 'above' and 'below' the GPS user.

Also, GPS is the name of the American GNSS (Global Navigation Satellite System) system.They are not the only ones out there. There is GLONASS, run by the Russians, Galileo which is run by the European Space Agency, Beidou 2 which is Chinese. I think Japan has it's own too, but it is not global, only for Japan. Nowadays, the GPS chips in your phones will often connect to multiple systems, to get a more precises coordinate resolution. Surveyors have been using GPS combined with GLONASS for years.

Food for thought.

User avatar
landmeterbeuckx
V.I.P Member
V.I.P Member
Posts: 250
Joined: Tue May 01, 2012 5:19 pm
Full Name: Lieven Beuckx
Company Details: Studiebureau Beuckx
Company Position Title: Owner
Country: Belgium
Linkedin Profile: Yes

Re: Georeferencing (Sorry)

Post by landmeterbeuckx » Thu May 31, 2018 1:48 pm

i measure everything in gps with translation to a Lambert coordinate system that's used in Belgium.

If it has to do with property borders etc everything stays in this system and is also transferred to the client. Architects want to draw in a local system with on face aligned with an x or y axis. When i deliver stuff like this i keep my original project and the copy is rotated and translated. If any work has to be done later on like setting out or measure additional stuff, i can easily transform and insert in the already delivered system.

For accuracy as Scott already mentioned only gps through rtk or with a base can be used. All other gps readings are worthless if you want to use the data later for construction purposes. The height with gps is absolutely not reliable. Even with rtk one has to measure each control point with a total station or even better a digital level. You take one measurement as reference for z and refer the rest to this one. Differences with rtk of several cm's are not uncommon.

Gps has to be taken in open canopy and under different constellations to be sure to be used.

In scene i usually measure a few paper targets or the center of some prisms with a total station and import on the scans level as a cor. I rename the corresponding targets to the point number in the control file.

With a riegl it is easier, you just put the rtk on top and do not have to worry about anything else. I do take control points for control.

Surveying is control and control and control...

EDIT : if anyone is interested i want to make a small screen recording of how it's done in scene.
LSBbvba
Surveying services - 3D Laserscanning
Tel : +3211661621
www.landmeter-beuckx.be
info@landmeter-beuckx.be

Cranfill.ca
I have made <0 posts
I have made <0 posts
Posts: 6
Joined: Thu May 12, 2016 3:33 pm
Full Name: Cody Cranfill
Company Details: Construction Engineering
Company Position Title: Scan Technician
Country: USA
Linkedin Profile: No

Re: Georeferencing (Sorry)

Post by Cranfill.ca » Thu May 31, 2018 2:23 pm

EDIT : if anyone is interested i want to make a small screen recording of how it's done in scene.

I would be very interested in seeing how you preform this in Scene.

Everyone around my office seems to do this differently.

User avatar
ddustin
V.I.P Member
V.I.P Member
Posts: 931
Joined: Mon Aug 22, 2011 9:28 pm
Full Name: David Dustin
Company Details: FARO Technologies Inc
Company Position Title: Global Dir of PS Product Management
Country: USA
Linkedin Profile: Yes
Location: Georgia, USA.. Planet Earth

Re: Georeferencing (Sorry)

Post by ddustin » Thu May 31, 2018 2:28 pm

landmeterbeuckx wrote:
Thu May 31, 2018 1:48 pm
EDIT : if anyone is interested i want to make a small screen recording of how it's done in scene.
Lieven,
I am interested.

Thank you all for the insight
David
Global Director of Public Safety - Product Management

User avatar
3DForensics
Global Moderator
Global Moderator
Posts: 1797
Joined: Mon Aug 03, 2009 1:52 am
Full Name: Eugene Liscio
Company Details: AI2-3D Forensics
Company Position Title: Owner
Skype Name: eliscio
Location: Toronto, Canada

Re: Georeferencing (Sorry)

Post by 3DForensics » Thu May 31, 2018 2:41 pm

A video on this would be great!

Eugene

Return to “FARO”