Help with targets for georeferencing

Discuss all Construction & Architecture related issues here.
Post Reply
Muaz
I have made 10-20 posts
I have made 10-20 posts
Posts: 19
Joined: Thu Mar 17, 2016 8:42 am
Full Name: Muaz Alzouby
Company Details: LTU
Company Position Title: Student
Country: Sweden

Help with targets for georeferencing

Post by Muaz » Tue May 03, 2016 9:43 am

Hey,

I am not a laser scanner guy and I have never scanned, however, Ill be scanning a rock tunnel for investigating the efficiency of 3d imaging for rock engineering purposes.

Now It is important that the data are geo referenced (some angles have to be calculated). The contractor of the tunnel sets up some ground reference points in the tunnel using a white spray making a plus mark (+)
Do you think its possible to georeference without a physical target ? a ball or something ?
If i also used the internal camera would it be easy to georeference and identify those ground points from the colored image?
or should I set up some kind of physical bodies as targets .. in this case any advice on some objects that can be used on the rock quickly ? like tennis balls ?

Its kind of urgent so the help is really appreciated

User avatar
Formula1982
V.I.P Member
V.I.P Member
Posts: 118
Joined: Tue Oct 13, 2015 4:02 pm
Full Name: Stephen Ferrari
Company Details: UTEC StarNet Geomatics
Company Position Title: Technical Lead
Country: United Kingdom

Re: Help with targets for georeferencing

Post by Formula1982 » Tue May 03, 2016 11:44 am

Hi.

Firstly, yes you can put physical objects (balls are good) in the scene and use these as reference targets. So long as they are clearly seen from adjacent scans. The scanner will probably pick up the white spray paint just fine in its intensity values, without the need to add photography.
It all depends what your tolerances are. A tunnel for example will likely have errors building up quite quickly. Adding a totalstation to the workflow to acquire targets would be best.

However given that you've never done laser scanning before, and what I assume is a difficult area to get back into if you mess up the first time, I would suggest you recruit an experienced surveyor to assist you with this project. It would be well worth the costs, especially if you later find that the data you generate yourself is not going to work.

I've done laser scans of tunnels for geological purposes before - identifying and mapping small faults and cracks in the rock structure on the walls and ceiling.

Muaz
I have made 10-20 posts
I have made 10-20 posts
Posts: 19
Joined: Thu Mar 17, 2016 8:42 am
Full Name: Muaz Alzouby
Company Details: LTU
Company Position Title: Student
Country: Sweden

Re: Help with targets for georeferencing

Post by Muaz » Tue May 03, 2016 12:55 pm

Hey Formula1982 and thanks for your quick reply!!

Execuse my ignorance, but lets say I used the marked fixed points that exist in the tunnel allready. If the centre of the + mark was kind of big , I understand that this will mean a considerable margin of error in geo referencing, right ? but the precision of the point cloud wont be affected right ?

Shouldn't I expect a a low accuracy, high precision systematic error in this case ?

In other words, If I want to for ex. calculate the dip of a crack in the rock .. I will still get an accurate angle, but the location of the crack might not be so accurate ?

The problem is the time of the test is very limited since the production must keep on going .. and am not sure If I have time or accessibiltiy to use a total station or set up better reference targets.

Am using P40 if mentioning that makes any difference

Rikore
V.I.P Member
V.I.P Member
Posts: 309
Joined: Sun Aug 16, 2015 6:48 pm
Full Name: Bruce Yager
Company Details: Rikore Geomatics
Company Position Title: Department Head
Country: USA
Linkedin Profile: No

Re: Help with targets for georeferencing

Post by Rikore » Tue May 03, 2016 1:23 pm

Mauz,
If you are using the P40, I would suggest strongly you use the traverse mode in the tunnel. Two targets, one for backsight and one for the foresight. Set a mark on the tunnel floor, could be a pencil "x" or the white paint with a dot to set over. This will give you great results as the P40 is excellent in the traverse mode.

Muaz
I have made 10-20 posts
I have made 10-20 posts
Posts: 19
Joined: Thu Mar 17, 2016 8:42 am
Full Name: Muaz Alzouby
Company Details: LTU
Company Position Title: Student
Country: Sweden

Re: Help with targets for georeferencing

Post by Muaz » Tue May 03, 2016 1:28 pm

Thanks alot Rikore,

Ill try to do that .. and btw is there any reference where i can read more about so called traverse mode ?

Rikore
V.I.P Member
V.I.P Member
Posts: 309
Joined: Sun Aug 16, 2015 6:48 pm
Full Name: Bruce Yager
Company Details: Rikore Geomatics
Company Position Title: Department Head
Country: USA
Linkedin Profile: No

Re: Help with targets for georeferencing

Post by Rikore » Tue May 03, 2016 2:33 pm

Mauz,
The field manual talks a bit about the traverse mode. I would also search Leica's YouTube area for videos on this. It is pretty easy. Pick traverse mode for setup, scan the target backsight and then the foresight. You can use any coordinates for a starting point and give the backsight target an azimuth. The field manual should come with the P40. You could get it also from Leica MyWorld site.

Bill-Wallace
V.I.P Member
V.I.P Member
Posts: 140
Joined: Thu Oct 29, 2009 11:14 pm
Full Name: Bill Wallace
Company Details: Leica Geosystems
Company Position Title: HDS Sr. Software Product Manager

Re: Help with targets for georeferencing

Post by Bill-Wallace » Tue May 03, 2016 3:45 pm

Mauz,

Here is some basic surveying info that you may or may not know but I always like to start at zero.

If you have marks on the ground you can setup a survey target pole over that mark. It will have a level on it so you can get it straight above the mark. The pole usually has sort of a tripod also. Some poles are adjustable in height and some have a fixed height. You make a note of the actual target height above the mark on the ground and then when you are entering info in Cyclone or on the scanner, you can enter the target height so Cyclone knows where it is on the ground.

You can use various types of targets on the pole.

Traversing is a tried and true surveying method. The scanner can also be setup directly over a point on the ground. The bracket (Tri-brach) that attaches the scanner to the tripod can have an eyesite on the side that lets you look directly down and line up exactly over the ground point. The scanner is then leveled and we also have a tilt compensator to ensure the data is collected level (critical for traverse). There is a special way to measure the scanner height and even a special tape measure (not required) to help get the scanner height. You need to add that into the system just like a target height.

With traversing you setup the scanner over one ground point and you set a target up over the next ground point. You scan the scene and you scan the "Foresight" target. This is the next place you will set the scanner.

Then you move the scanner to the next ground point (that you just scanned. And you also setup targets over the last ground point (where the scanner used to be) and another target over the next ground point (where the scanner will be later). You scan the scene. Then you scan the last point target (your backsight) and you scan the target at the next ground point (foresight). You continue this leap frogging down the line of ground points.

For the most accuracy you try to loop back to the beginning point (closed loop) but you don't have to do that and in a tunnel it may not be practical or best.

If you learn how to do a traverse using our scanners (nobody else does it that I know of) then when you import the data it is already registered. If you have a control file that tells you the geolocation of some of the ground points, then you are also georeferenced.

That is a very generic description. You'll have a bit of learning but maybe a local sales rep or local survey company doing scanning can be contracted to assist at first. Our training classes will surely cover traversing with a scanner as well. It can be kind of perfect for tunnels.

Muaz
I have made 10-20 posts
I have made 10-20 posts
Posts: 19
Joined: Thu Mar 17, 2016 8:42 am
Full Name: Muaz Alzouby
Company Details: LTU
Company Position Title: Student
Country: Sweden

Re: Help with targets for georeferencing

Post by Muaz » Tue May 10, 2016 12:09 pm

Hi Bill-Wallace,

Thanks for the details info, I did not know most of it !

I have allready scanned using targets .. there were allready some ground/reference points in the tunnel, so placing the targets on these points has not been so hard !

We have used 3 targets with known coordinates inserted into the scanner, and 1 more target on the ground with unknown coordinate ( what is the benefit gained from the last target ? )

I liked how the scanner checks the target coordinates, matches them and even estimates the error in 3 directions, any thoughts on the accuracy of this system ?

Even though there is alot for me to learn, I have to say I enjoyed my first scan :)

Oxbow
V.I.P Member
V.I.P Member
Posts: 252
Joined: Wed Aug 06, 2008 10:07 pm
Full Name: Eric Bergholz
Company Details: Laserscanning Europe
Company Position Title: CEO
Country: Germany
Skype Name: eboxbow
Linkedin Profile: Yes
Contact:

Re: Help with targets for georeferencing

Post by Oxbow » Fri May 13, 2016 10:09 am

Hi,

if you have reference points in the tunnel you can use reference sphere and miniprism to measure them in your grid systems. This video show`s you the workflow.


youtu.be/

for any question feel free to contact use or check our laserscanner accessories web shop.

..Eric

cpw006
V.I.P Member
V.I.P Member
Posts: 282
Joined: Tue Nov 20, 2007 3:51 am
Full Name: Chris Widmer
Company Details: LAS
Company Position Title: 3D Scanning Manager
Country: USA
Linkedin Profile: Yes
Location: Atlanta, Georgia

Re: Help with targets for georeferencing

Post by cpw006 » Fri Jul 29, 2016 3:40 pm

How did the tunnel project go? What method did you use? If you used the traverse function with the scanner, how well did it turn out?

Post Reply

Return to “Construction & Architecture”