Registration valid, telephone poles say different

Discuss all Cyclone Leica Cyclone, Field 360 & Register 360 related issues here.
Post Reply
Spwilka
I have made 10-20 posts
I have made 10-20 posts
Posts: 14
Joined: Wed Nov 06, 2019 7:45 pm
Full Name: Alison Wilkes
Company Details: NYSP
Company Position Title: Analyst
Country: USA
Linkedin Profile: No

Registration valid, telephone poles say different

Post by Spwilka »

Hello, I have a scene that I'm having issues with in Cyclone. I get a valid registration each time I try to register, but every time I do a visual check, by slicing or placing a limit box, the results are very off.

I sliced a telephone poll and instead of a perfect circle I am getting all sorts of scan overlap problems. Sometimes I'm getting a doubling of objects as well. (see below)
The area was scanned with a P40, and leveled. I had done a prior scene that day that registered perfectly.
I have many overlapping points between scan locations, and have tried an auto registry, a cloud to cloud with constraints using pick points, and the visual alignment. Each time the registration is valid, but the the edges are not clear.
I know there's a lot of possible reasons and I don't have all of the info on here, but does anything stick out as a reason why my modelspace isn't matching up well, even though the error rates are lower?
Object registration issue.jpg
Registration issue.png
You do not have the required permissions to view the files attached to this post.
Last edited by Spwilka on Mon Mar 30, 2020 8:41 pm, edited 2 times in total.

Rikore
V.I.P Member
V.I.P Member
Posts: 350
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
Been thanked: 6 times

Re: Registration valid, telephone poles say different

Post by Rikore »

Alison,
Since you have tried the different ways, visual, etc., I am thinking one of the scans may be off level. Try to start with just 3 or 4 scans, register and look at the results in the model space. If the scans look good, then add another scan. When the scans don't look good, then remove the last one and add all the others. Then look at the results. One of the scans could be causing the issue. Maybe the scanner went out of level after it started or had an error. I usually start with visual on the first two scans. I've had this happen several times when along a roadway. Wind, truck passing can vibrate the scanner.

User avatar
JLarson
V.I.P Member
V.I.P Member
Posts: 173
Joined: Mon Jun 14, 2010 4:57 pm
Full Name: Jason Larson
Company Details: NAIK Consulting Group
Company Position Title: Director of Survey NY
Country: US
Linkedin Profile: Yes
Location: New York City

Re: Registration valid, telephone poles say different

Post by JLarson »

Another potential issue could be vegetation. If you have a lot of trees in the area, the cloud to cloud alignments can be massively off (but still show good results). You may want to trim the tress out of the registration and then re-run the cloud alignments and see if that makes a difference.
Jason Larson, PLS
NAIK Consulting Group
New York, NY

User avatar
steves01x
V.I.P Member
V.I.P Member
Posts: 357
Joined: Tue Jul 21, 2015 3:58 pm
Full Name: Steve Smith
Company Details: Scanning
Company Position Title: Surveyor
Country: Scotland
Linkedin Profile: Yes
Has thanked: 16 times
Been thanked: 17 times

Re: Registration valid, telephone poles say different

Post by steves01x »

As you only have 9 scans it wont take long, go through each pair of scans with the visual align, re-do them and it should sort it. There is also a chance that in the registration results the one with the highest group error (much higher than the rest) is your problem that is knocking the rest of the scans wrong.

MajorDomo
V.I.P Member
V.I.P Member
Posts: 145
Joined: Mon Feb 25, 2019 4:42 am
Full Name: Major Domo
Company Details: VeritasVfx
Company Position Title: ceo
Country: Portugal
Linkedin Profile: No
Has thanked: 7 times
Been thanked: 27 times

Re: Registration valid, telephone poles say different

Post by MajorDomo »

Delete moving stuff and reflections from your control space?

Was the tilt compensator turned on? are the scans level? Has someone banged the scanner? Did you create new stations on each setup, or are have you got a few scans on the same scanworld, when they should be split?

Tookie
V.I.P Member
V.I.P Member
Posts: 101
Joined: Thu Sep 21, 2017 9:43 pm
Full Name: Iain Took
Company Details: WMP
Company Position Title: Dogsbody
Country: United Kingdom
Linkedin Profile: No
Has thanked: 1 time
Been thanked: 5 times

Re: Registration valid, telephone poles say different

Post by Tookie »

I would set all the weights to 1 and cut any trees from the control spaces. Looks like a fairly sparse scene, what is the spacing between scans? Do you have low overlap?

I normally look for the link with the biggest error and start there.

User avatar
steves01x
V.I.P Member
V.I.P Member
Posts: 357
Joined: Tue Jul 21, 2015 3:58 pm
Full Name: Steve Smith
Company Details: Scanning
Company Position Title: Surveyor
Country: Scotland
Linkedin Profile: Yes
Has thanked: 16 times
Been thanked: 17 times

Re: Registration valid, telephone poles say different

Post by steves01x »

Where is scan 6?

Also in modelspace, click the 2 poles to find out which scans they belong to and c2c them to correct it.

Post Reply

Return to “Leica Cyclone, Field 360 & Register 360”