Cyclone vs Register 360

Discuss Leica Geosystems hardware here.
Post Reply
Amanda
I have made <0 posts
I have made <0 posts
Posts: 6
Joined: Wed Jan 20, 2021 9:47 pm
3
Full Name: Amanda Renn
Company Details: Geomatics
Company Position Title: Technologist
Country: Canada
Has thanked: 1 time

Cyclone vs Register 360

Post by Amanda »

Hi everyone,

I'm fairly new to laser scanning and I would like to figure out when to use which software. I know this often depends on the individual project, but I'm trying to get a general idea when people choose which software - and why.

Most of our projects have between 100 and 300 scan setups. We use Leica P-Series scanners.

Using auto cloud and/or smart align in Register 360 seems to be a great option because it saves a lot of time not having to register the scans together manually. However, I found that the registration result isn't always great. There is usually an abundance of links created and during the usual QC I have found some badly misaligned scans in the bundle - despite a great bundle error report. Both required manual, time consuming clean-up - which makes me think using Cyclone to begin with would have taken similar time. Has anyone had a similar experience?

Thank you for any input.
Jamesrye
V.I.P Member
V.I.P Member
Posts: 643
Joined: Mon Aug 11, 2008 4:13 pm
15
Full Name: James Rye
Company Details: Merrett Survey Partnership
Company Position Title: Spatial Analyst
Has thanked: 28 times
Been thanked: 69 times

Re: Cyclone vs Register 360

Post by Jamesrye »

Use targets, or at least have independent control. Cloud to Cloud can give great looking stats, but can be completely misaligned.

Don't fall for marketing hype :lol:
mstachoni
I have made 30-40 posts
I have made 30-40 posts
Posts: 32
Joined: Wed Oct 17, 2018 1:49 pm
5
Full Name: Matt Stachoni
Company Details: Tutor Perini Corporation
Company Position Title: BIM Manager
Country: USA
Linkedin Profile: Yes
Been thanked: 20 times
Contact:

Re: Cyclone vs Register 360

Post by mstachoni »

Many people including myself use both.

Cyclone Register is the Unix of laser scanning registration software. It's old, creaky, has a UI straight out of 1991, and almost impossible to learn adequately on your own. It's the most opaque and hard to use piece of software I've ever dealt with, but if you get proper training it's powerful as all get out. There's literally almost nothing it can do, except make you feel comfortable in using it.

Reg 360 is 180° different from Cyclone Register. It's easy to self-learn (MyWorld has a great course on it), is very fast, and has enough power to get your scans registered. It hides the muddy verbiage used in Cyclone and the UI is easy to understand. That said, there are things that Cyclone Register does better, and there are a few things you can only do in Cyclone, particularly on larger more complex projects. But for many people that "better" may not be worth the pain of learning it. If you are regularly working with 100+ scans per daily project output you may need it. Many people including myself will import and register in Reg 360 then take it into Cyclone Register to tighten up the registration.

I typically forego using any auto-registration methods, it's just not worth the time in my use cases. But that is specific to the project. I've had scanning sessions where the auto-registration routine in Reg 360 was amazing and saved hours of time. I think it depends on how different each scan is; scans full of repetitive objects, such as a structural framing system under construction, or a long featureless hallway, will send the auto-registration algorithm on a wild goose chase and nothing will align properly.
- Matt
fpensa
I have made 50-60 posts
I have made 50-60 posts
Posts: 52
Joined: Wed Jan 24, 2018 12:51 pm
6
Full Name: Gianni Epinotto
Company Details: GC CM
Company Position Title: BIM Manager
Country: Canada
Has thanked: 9 times
Been thanked: 9 times

Re: Cyclone vs Register 360

Post by fpensa »

We use the preregistration method with Cyclone 360 but never trust it... It's good to have scans roughly in the right position tough, this speed up the quality control (mostly using the slicer) and allow to quickly correct misalignments deleting problematics links and proceeding to a manual alignement and a Cloud to Cloud refining, trying to proceed per bundles so Cyclone Register also can proof the registration proposing extra links. We rarely need to use targets (square or spheres), mostly when we scan flat slabs with almost no vertical elements or when we need to georeference the cloud
Amanda
I have made <0 posts
I have made <0 posts
Posts: 6
Joined: Wed Jan 20, 2021 9:47 pm
3
Full Name: Amanda Renn
Company Details: Geomatics
Company Position Title: Technologist
Country: Canada
Has thanked: 1 time

Re: Cyclone vs Register 360

Post by Amanda »

Thank you for your input!
Inception
I have made 10-20 posts
I have made 10-20 posts
Posts: 13
Joined: Wed Feb 17, 2021 1:53 am
3
Full Name: Travis Long
Company Details: CADDS Group
Company Position Title: Surveyor
Country: Australia
Linkedin Profile: No
Has thanked: 4 times

Re: Cyclone vs Register 360

Post by Inception »

I haven't ever used Reg360, but a couple of the guys in my office have used it in the past.

They look at it like this:

Reg360 is great software for small scan areas with nice surfaces to register off and a small total scan count. Cyclone is a much more numbers driven process and requires a bit more experience to get a clean reg - however, it can provide better results (especially in more complex environments). You need to know how/where to look for bad links via slicing/limit boxing; and with this, have an idea of where the error is coming from.

Reg360 is the simpler, less powerful version of Cyclone that can give you results quicker in some scenarios. We now pretty much exclusively use Cyclone as it's just a more powerful & versatile platform. In your case where projects range from 100-300 scans, I think you'd be crazy to not be using Cyclone.
Last edited by Inception on Wed Mar 31, 2021 6:03 am, edited 1 time in total.
Inception
I have made 10-20 posts
I have made 10-20 posts
Posts: 13
Joined: Wed Feb 17, 2021 1:53 am
3
Full Name: Travis Long
Company Details: CADDS Group
Company Position Title: Surveyor
Country: Australia
Linkedin Profile: No
Has thanked: 4 times

Re: Cyclone vs Register 360

Post by Inception »

Jamesrye wrote: Mon Mar 15, 2021 7:01 pm Use targets, or at least have independent control. Cloud to Cloud can give great looking stats, but can be completely misaligned.

Don't fall for marketing hype :lol:
C2C can certainly give strong statistical results but have a dodgy reg - that's just a part of intermittent checking between steps because algorithms are never perfect. As I mentioned above, it does take some experience to learn how to properly error check & troubleshoot for these reg errors in Cyclone.

Not sure about it being 'marketing hype'... We hardly ever use targets for our RTC360 reg, always C2C (except for a few special scenarios). We have completed lots of large scale projects (~500m scan areas) in complex environments and fitted the C2C-only reg directly on to our control to within ~10-15mm. Good operators/knowledge = good output with C2C in Cyclone ;)
User avatar
Mike Annear
V.I.P Member
V.I.P Member
Posts: 800
Joined: Mon Jun 08, 2009 5:45 am
14
Full Name: Mike Annear
Company Details: The Yard Brisbane
Company Position Title: Manager - 3D Design and Scanning
Country: Australia
Linkedin Profile: Yes
Location: Hemmant, Queensland, Australia
Has thanked: 122 times
Been thanked: 196 times
Contact:

Re: Cyclone vs Register 360

Post by Mike Annear »

Inception wrote: Wed Mar 31, 2021 6:01 am
Jamesrye wrote: Mon Mar 15, 2021 7:01 pm Use targets, or at least have independent control. Cloud to Cloud can give great looking stats, but can be completely misaligned.

Don't fall for marketing hype :lol:
~10-15mm. Good operators/knowledge = good output with C2C in Cyclone ;)

10-15 mm = Good ??
Mike Annear
www.mikeannear.com
QLD, Australia.
Jamesrye
V.I.P Member
V.I.P Member
Posts: 643
Joined: Mon Aug 11, 2008 4:13 pm
15
Full Name: James Rye
Company Details: Merrett Survey Partnership
Company Position Title: Spatial Analyst
Has thanked: 28 times
Been thanked: 69 times

Re: Cyclone vs Register 360

Post by Jamesrye »

Mike Annear wrote: Wed Mar 31, 2021 8:16 am
Inception wrote: Wed Mar 31, 2021 6:01 am
Jamesrye wrote: Mon Mar 15, 2021 7:01 pm Use targets, or at least have independent control. Cloud to Cloud can give great looking stats, but can be completely misaligned.

Don't fall for marketing hype :lol:
~10-15mm. Good operators/knowledge = good output with C2C in Cyclone ;)

10-15 mm = Good ??

For cloud registration, 10-15mm RMS with a BLK is 'good' - at least it matches with my experience - see C2C stats below. But you can get this sort of result and yet have two clouds that are misaligned, you always have to check the result visually. Hence why we always have some targets in our scans - for redundancy. For anyone interested, here's a post from way back, looking at my own tests on cloud registration accuracy:
https://laserscanningforum.com/forum/vi ... php?t=8214

Cloud/Mesh 1 [Station-004: SW-004 (Leveled) : Station-005: SW-005 (Leveled)]
Translation: (2.634, 39.843, 0.669) m
Rotation: (-0.0001, 0.0001, -1.0000):-96.889 deg

Objective Function Value: 0.000144748 sq m
Iterations: 100
Overlap Point Count: 359431
Overlap Error Statistics
RMS: 0.0145836 m
AVG: 0.0107217 m
MIN: 5.47117e-09 m
MAX: 0.0497937 m
Overlap Center: (7.287, 19.692, 6.709) m
Filter Parameters
Max Search Distance: 0.100 m
No Subsampling.
Stopping Criteria
Max Iterations: 100
User avatar
stutosney
V.I.P Member
V.I.P Member
Posts: 239
Joined: Tue May 15, 2018 5:55 am
5
Full Name: Stuart Tosney
Company Details: ---
Company Position Title: Associate Director - Survey
Country: London
Linkedin Profile: Yes
Has thanked: 25 times
Been thanked: 58 times

Re: Cyclone vs Register 360

Post by stutosney »

Jamesrye wrote: Wed Mar 31, 2021 9:11 am
For cloud registration, 10-15mm RMS with a BLK is 'good' - at least it matches with my experience - see C2C stats below. But you can get this sort of result and yet have two clouds that are misaligned, you always have to check the result visually. Hence why we always have some targets in our scans - for redundancy. For anyone interested, here's a post from way back, looking at my own tests on cloud registration accuracy:
https://laserscanningforum.com/forum/vi ... php?t=8214

Would be good to do a modern day comparison using the latest version and sample with smart align and auto-align and then compare
Post Reply

Return to “Leica Geosystems”