Faro quality setting

Faro S150, S350, X30, X130, X130HDR, X330, X330HDR, M70, Trimble X7, Leica RTC360, HDS6*00, HDS7000, Z+F IMAGER 5016, 5010X, 5010C, 5006H, 5006EX & PROFILER 9012 etc
User avatar
jcoco3
Global Moderator
Global Moderator
Posts: 1493
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: No
Has thanked: 29 times
Been thanked: 58 times
Contact:

Re: Faro quality setting

Post by jcoco3 »

Thanks Si!

I guess the next thing to think about is the difference in quality between the X series and the Focus 120. I know the X series was advertised as "half" the noise of the 120, but specifically what was improoved still aludes me. Was the range error reduced maybe as a function of the new wavelength? Was there an improovment in angular accuracy, or was there just some changes to the internal noise compression filters, or possibly a little improovment to all?

What I am still trying to establish, is if you use an X series at a lower quality setting for speed, could you effectivley achieve similar results as a 120 at a higher setting? I am a bit of a speed freak sometimes and while quality is important, we are not having any problems with the noise typical of a Focus 120 scan.

I am not directing these question to you Si as I know some of this could be proprietary...just expressing my curiosities to potentially exploit an under appreciated advantage of the X series. More reasons to conduct some experiments :ugeek:

adriantaylor09
I have made 60-70 posts
I have made 60-70 posts
Posts: 68
Joined: Wed May 07, 2014 8:57 pm
Full Name: Adrian Taylor
Company Details: Revisia Ltd
Company Position Title: Industrial 3D Specialist
Country: New Zealand
Linkedin Profile: Yes

Re: Faro quality setting

Post by adriantaylor09 »

Thanks for the answers guys.. very helpful. It's always interesting to learn what settings people are running with.
brianpgreen wrote:I think you'll find the general consensus is that you won't see much difference between 3x and 4x.
I'll definitely be testing this theory out! using 3x would save me a lot of time, and from the sounds of things, I won't be trading off too much in terms of accuracy.

User avatar
3DForensics
Global Moderator
Global Moderator
Posts: 1909
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
Has thanked: 1 time
Been thanked: 7 times
Contact:

Re: Faro quality setting

Post by 3DForensics »

Dedken wrote:
jcoco3 wrote:
All that being said, I am not sure the quality setting really affects accuracy as much as it just reduces the ambient noise.
I would go along with that, it's pretty much what the Focus manual says straight out. It's more about noise reduction.
I was reading this over and in my mind, more noise is less accurate when looking at an individual point, especially if you are taking less samples. If this were a total station taking 3,000 samples, then I would expect greater accuracy on distance measurements.

However, I agree that if this is a type of random noise and you try to fit a sphere or plane, then it may not be easily distinguishable.

User avatar
mckevers
I have made <0 posts
I have made <0 posts
Posts: 1
Joined: Thu Jul 18, 2013 3:34 pm
Full Name: Scottt McKever
Company Details: Canyon Logics 3D Laser Scanning to BIM
Company Position Title: Owner
Country: Canada
Linkedin Profile: Yes

Re: Faro quality setting

Post by mckevers »

Jonathan, because you are comparing the two products I thought you might be interested in the results of a test I did to determine whether the light frequency of the X series would pick up more points on black surfaces than my 120. See the comparison images attached.

While the two sets of scans were done at two different times, introducing the possibility that natural lighting factors may have played a part in the results, the light conditions and placement of the scanners were very similar.
You do not have the required permissions to view the files attached to this post.

User avatar
jcoco3
Global Moderator
Global Moderator
Posts: 1493
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: No
Has thanked: 29 times
Been thanked: 58 times
Contact:

Re: Faro quality setting

Post by jcoco3 »

Thanks Scott, I have seen similar results before with scans of black cars and I have even done a little testing of my own on a battleship deck. I wouldn't catagorize this difference as a difference in noise or quality, but there is no doubt the X series has no problem in a competition with the 120 in bright daylight using the new wavelength. It is a night and a day difference on subject you scanned :)


Edit:
Oh by the way, welcome to the forum Scott :)
Last edited by jcoco3 on Mon Feb 16, 2015 9:25 am, edited 1 time in total.

User avatar
jcoco3
Global Moderator
Global Moderator
Posts: 1493
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: No
Has thanked: 29 times
Been thanked: 58 times
Contact:

Re: Faro quality setting

Post by jcoco3 »

I was reading this over and in my mind, more noise is less accurate when looking at an individual point, especially if you are taking less samples. If this were a total station taking 3,000 samples, then I would expect greater accuracy on distance measurements.
Eugene, I follow your logic, and this sounds very reasonable, but I wonder if on a sphere with a surface designed to be a good target for scanning if the difference in range noise is even noticeable between a low or high quality setting? Yet another good reason for an experiment :P

I may setup and run the experiment this week, but I had hope that somebody else had already put the effort in and was willing to share.

User avatar
3DForensics
Global Moderator
Global Moderator
Posts: 1909
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
Has thanked: 1 time
Been thanked: 7 times
Contact:

Re: Faro quality setting

Post by 3DForensics »

Very nice example between the two scanners. It's quite a drastic difference.


Sent from my iPhone using Tapatalk

User avatar
jcoco3
Global Moderator
Global Moderator
Posts: 1493
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: No
Has thanked: 29 times
Been thanked: 58 times
Contact:

Re: Faro quality setting

Post by jcoco3 »

So we finally had time to run the quality comparison test scans, but we botched it :( Well it was not a total loss. I naively assumed that if the scanner remained in the same place that the horizontal zero start position would be the same for a series of repetitive scans set to various quality settings. Obviously there has to be some small amount of angular error in the base servo or angular encoder so the horizontal zero position has small angular differences, as indicated by the rotations being slightly off in the transformation dialog of each scan in Scene. Silly mistake, and I should have known better. At first I thought the error might have been due to the compass, but we turned off all the sensors except for the inclinometer before conducting the test. We even placed some checkerboards around the scanner position to tie to a total station for back-checking our scans, but because their poor placement I don't think they are sufficient to register the scans together with enough angular accuracy to properly align the scans for comparison. We will try re-run the procedure using a different methodology soon, but until then I wanted to share what was salvageable from the results.

Here are some images of spheres scanned at 1x and 4x quality setting with spheres placed at 10', 30' and 50'
120S_1X_FRONT.png
120S_4X_FRONT.png
As the quality setting is increased the positioning of each point appears to be slightly more organized the range noise also appears to be less random. All sphere fits were automatic/green and rotating the scans to nearly the same angular position the difference in range to each sphere at all quality settings represented less than 1mm of variance, although I am not satisfied with this result due to our inability to nail down the exact horizontal orientation. I think in the next test we will setup the same 3 test spheres, but we will also surround the scanner with many spheres at many different heights and ranges to be used to registered the scans together while ignoring the 3 test spheres. After registration we will re-run the automatic sphere search to find the 3 test spheres and examine the results.

Initial conclusions:
1. Higher quality settings capture more points on a spheres at longer ranges, but I think we all knew that already.
2. As spheres are placed further away from the scanner the weighted importance of each point should increase as the number of acquired points decrease, therefore higher noise overall should negatively effect the sphere fit, but by measuring the distance between spheres the initial results don't suggest a large impact as the spheres from different quality settings all appear to be near the same position(less than 1mm deviation). In short it would seem like even with 85 noisy points on a sphere an accurate sphere fit can still be established.

I plan to provide some tabulated results in excel comparing sphere positions from one quality setting to another and the sphere positions to total station shots after we re-run the experiment. I would also like to compare scanned sphere point coordinates to the scanner's origin in an attempt to asses the range noise per quality setting. This is turning out to be more difficult than I anticipated, but I am confident we can obtain some clear results that are easy to understand in time :)

Full resolution images here:
http://1drv.ms/1yWYCfD
You do not have the required permissions to view the files attached to this post.

User avatar
3DForensics
Global Moderator
Global Moderator
Posts: 1909
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
Has thanked: 1 time
Been thanked: 7 times
Contact:

Re: Faro quality setting

Post by 3DForensics »

You can see a noticeable difference between 1X and 4X, but certainly not anything alarming or unexpected.

Thanks for the high resolution images!

User avatar
Dedken
V.I.P Member
V.I.P Member
Posts: 363
Joined: Fri Mar 15, 2013 10:28 am
Full Name: Kenneth Bazley
Company Details: Sir Robert McAlpine
Company Position Title: Senior Geospatial Engineer for HDS
Country: UK
Linkedin Profile: Yes
Location: London

Re: Faro quality setting

Post by Dedken »

Nice work Jonathan - Though, I really don't know how you find the time to do all this as well as your day-to-day work!
All views are my own and are not representative of my employer, The Queen, God or anyone else for that matter.

"we need an instrument, to take a measurement" - I.MacKaye 1992

Post Reply

Return to “Phase Based Scanners”