Scene - target based registration - influence of inklinometer

Discuss all Faro related issues here.
User avatar
Global Moderator
Global Moderator
Posts: 1325
Joined: Sun Mar 04, 2012 5:43 pm
Full Name: Jonathan Coco
Company Details: Forte and Tablada
Company Position Title: AMM Division Leader
Country: USA
Linkedin Profile: Yes

Re: Scene - target based registration - influence of inklinometer

Post by jcoco3 » Fri Oct 13, 2017 6:31 pm

Sorry for the delayed response, been very busy and this topic requires focus.
Question is how does SCENE treats multiple inclinometr readings? Does it pick first one? That one which has lowest error? Or does it take some kind of average? Average or better median would be very useful in this case because it would eliminate random errors.
So, if my memory serves me (don't trust my memory, search the forum instead) at Scene version 5.3 there was change to allow some sort of inclinometer averaging as prior to this the reference scan's inclinometer reading was used to level the entire cluster or project. If the reference scan had a "bad level" then you could have a fairly significant level issue depending upon the size of your project and other variables. To the best of my knowledge the averaging approach is what is used for all types of registration from version 5.3 and beyond. There was a lengthy thread related to this issue that I cannot seem to find. I believe Brian Green was trying to figure out how the new averaging method worked, but I don't know if he found what he was looking for :?
I am not sure if I understand right but are you saying that relative correspondences are not able to tilt scan from its measured attitude (by its inclination sensor) during the scan registration no matter whether inclinometr is on or off?
No, sorry for the confusion. What I meant was, If the inclinometer is off then spheres and checkers or even cloud to cloud should tilt the scans, but if it is on then the only thing that can affect the tilt is external references.
And wouldn't it be illogical if relative correspondences wouldn't have power to slightly tilt the scan to achieve better alignment of the correspondences?
Hehe :D , I was certainly not expressing any agreement/disagreement with the logic, just stating knowledge that was passed to me that I consider factual...for now. I think there are some good reasons, and maybe some not so good reasons but I have never fully analyzed which situation is better. I have given some thought on how to setup a method for testing this information in Scene, but I have yet to make an attempt.
I did a small test and I compared global coordinates of one point and there is significant difference in height between both approaches. So I don't understand what you mean.. Please see attached files to see related scan managers.
Yikes! You might be on to something :o I think the information I was provided or my interpretation of it could be wrong. I just did a very similar test with just three scans registered(cloud to cloud with inclinometer on) together than locked and placed withing a another cluster prior to target based registration to some GPS control points. I did one registration(in v5.3) with inclinometer ticked on in the general tab of the place scans dialog, and another attempt with the inclinometer ticked off...2 different results :x There was basically a 1/2" difference between one of three points in the reference scan and its corresponding GPS control point. The other 2 were off by 0.4" and 0.05".

I am sorry if I misguided you, I think I was wrong. Maybe the better interpretation of what I was told is that external references are the only thing that can "influence"(as opposed to tilt) as scan or project as it would appear as if Scene 5.3 is still heavily relying on inclinometer data. I don't know if it is the same for 7.0, but would think so. What version are you using? Also, way back in early versions you were supposed to be able to add a 0 or 1 to the last column of the csv to indicate the quality of the external reference point. I wonder if that would have still have an effect today?

Some supporting images and information for what its worth.

Transformation mismatch:
No inclinometer with GPS control 0.368898°
With inclinometer and GPS control 0.001989°
Placed by sensors using inclinometer 0°

You can't really see the difference below but if you download and toggle the images back and forth the tilt changes between the two methods.
without inclionmeter.jpg
with inclionmeter.jpg
This Sucks! I have been doing this steady for 6 years now and I am still making mistakes and re-learning things I thought I knew. Oh well, it is the end of the week for me, time to give it a rest and re-group on Monday.
You do not have the required permissions to view the files attached to this post.

Return to “Faro”