Deformation analysis statistics

Discuss Leica Cyclone 3DR software here.
Post Reply
StuEv
I have made 50-60 posts
I have made 50-60 posts
Posts: 52
Joined: Mon Nov 16, 2015 12:52 am
8
Full Name: Julian
Company Details: University
Company Position Title: Student
Country: Germany
Linkedin Profile: No

Deformation analysis statistics

Post by StuEv »

Hello,

i am doing a deformation analysis between a mesh and a point cloud. As a result i am getting a histogram with the results. I did this between two epoches on a wall with HDS7000 to analyse the scanners accuracy. How would you guys interpretate the accuracy of HDS 7000 in this picture?
Is it possible to get statistical results like standard deviation or a deviation distribution? Here is a screenshot of the result:
Reshaper2.png
@Gilles (3DReshaper): i got a student license from your company for this programm but i dont habve full access to the inspection tools. Is it possible to get a full access student license because it is used for my bachelor's degree?

Thank you

StuEv
You do not have the required permissions to view the files attached to this post.
User avatar
3DForensics
Honorary Member
Honorary Member
Posts: 1979
Joined: Mon Aug 03, 2009 1:52 am
14
Full Name: Eugene Liscio
Company Details: AI2-3D Forensics
Company Position Title: Owner
Skype Name: eliscio
Location: Toronto, Canada
Has thanked: 13 times
Been thanked: 70 times
Contact:

Re: Deformation analysis statistics

Post by 3DForensics »

What is the reference mesh? Is it a converted point cloud or a CAD model?


Sent from my iPad using Tapatalk
StuEv
I have made 50-60 posts
I have made 50-60 posts
Posts: 52
Joined: Mon Nov 16, 2015 12:52 am
8
Full Name: Julian
Company Details: University
Company Position Title: Student
Country: Germany
Linkedin Profile: No

Re: Deformation analysis statistics

Post by StuEv »

it is a mesh of a point cloud.

stuev
User avatar
3DForensics
Honorary Member
Honorary Member
Posts: 1979
Joined: Mon Aug 03, 2009 1:52 am
14
Full Name: Eugene Liscio
Company Details: AI2-3D Forensics
Company Position Title: Owner
Skype Name: eliscio
Location: Toronto, Canada
Has thanked: 13 times
Been thanked: 70 times
Contact:

Re: Deformation analysis statistics

Post by 3DForensics »

Only reason I ask is that if there is any simplification of the points/mesh and you are looking for fine results, then it could be a little difficult to interpret.
StuEv
I have made 50-60 posts
I have made 50-60 posts
Posts: 52
Joined: Mon Nov 16, 2015 12:52 am
8
Full Name: Julian
Company Details: University
Company Position Title: Student
Country: Germany
Linkedin Profile: No

Re: Deformation analysis statistics

Post by StuEv »

hmm what would you suggest to make a good mesh in 3dreshaper?
i have a point cloud density of 1cm. i made the mesh with 2cm mean space between points and hole detection of 6cm.after that i interpolatet existing points to refine the mesh.
I want to compare the mesh with a point cloud to investigate the accuracy of hds 7000 scans.
User avatar
3DForensics
Honorary Member
Honorary Member
Posts: 1979
Joined: Mon Aug 03, 2009 1:52 am
14
Full Name: Eugene Liscio
Company Details: AI2-3D Forensics
Company Position Title: Owner
Skype Name: eliscio
Location: Toronto, Canada
Has thanked: 13 times
Been thanked: 70 times
Contact:

Re: Deformation analysis statistics

Post by 3DForensics »

Can you try cloud to cloud distances in CloudCompare and see how that compares to your mesh to cloud result?

I would keep the original point cloud densities as they are and no need to decimate.

Eugene
User avatar
gilles_3DR
V.I.P Member
V.I.P Member
Posts: 235
Joined: Mon Dec 08, 2014 11:41 am
9
Full Name: Gilles Monnier
Company Details: Hexagon
Company Position Title: 3DR Technical Manager
Country: France
Linkedin Profile: Yes
Has thanked: 2 times
Been thanked: 28 times

Re: Deformation analysis statistics

Post by gilles_3DR »

Hi StuEv,

yes, for sure, you can request for a full temporary license to get all the inspection tools you need for this study.
We'd be happy to help.

In general, I would assume that comparing a cloud to a mesh is more precise than comparing a cloud to another cloud as the projection of a point onto a cloud is more likely to be method dependent.
The main reason to me is that the mesh you will create in Reshaper will be located in the middle of the noise thickness! Creating a good mesh for this application will therefore depend on the point density and an estimated value of scanner accuracy! If you need more help on how to create a good mesh for this application in Reshaper, do not hesitate to ask!
Anyhow, the cloud to cloud inspection will be available in Reshaper in the 2016 version which will most likely be available as a beta version in 4-5 weeks (just finalized implementation and currently running quality checks)!
This way, the 2 methods will be available for comparison.

About statistics distribution, in the image you shared, you can see the percentage of point within the deviation values. This is what I would call statistics distribution... What else would you expect?
About standard distribution, I would suggest to export your point cloud to csv and then evaluate the standard deviation in excel... (looks like you are showing off a missing functionality in Reshaper!)
StuEv
I have made 50-60 posts
I have made 50-60 posts
Posts: 52
Joined: Mon Nov 16, 2015 12:52 am
8
Full Name: Julian
Company Details: University
Company Position Title: Student
Country: Germany
Linkedin Profile: No

Re: Deformation analysis statistics

Post by StuEv »

Hey Gilles,
thanks for your reply!
Well i have one Point cloud which i thinned out by unifying it in Leica cyclone with a mean space of 2cm.
After Unifying, i still have 1cm point spacing and thats what i want.
Now to create a mesh for deformation analysis, i choose the Mesh in two steps method. In the first Step i selected a mean space between meshed points of 2cm and filling holes with 6cm. I did 2cm because if choose 1cm i get many holes in my point cloud :(. in the next step i refine my mesh with points of the point cloud and after that i smooth out the mesh. For the second mesh i have a bad spacing because the point cloud has a spacing in the middle of 6mm and in left and right direction up to 60mm!!! what would you suggest to do with it?

I already asked for a full license for 3dreshaper but i only get the automatic answering with the student license.
I also sent the filled form to get the full license :/.

Thank you all for your great help

StuEv
User avatar
gilles_3DR
V.I.P Member
V.I.P Member
Posts: 235
Joined: Mon Dec 08, 2014 11:41 am
9
Full Name: Gilles Monnier
Company Details: Hexagon
Company Position Title: 3DR Technical Manager
Country: France
Linkedin Profile: Yes
Has thanked: 2 times
Been thanked: 28 times

Re: Deformation analysis statistics

Post by gilles_3DR »

with 1cm point spacing, what is the size of your point cloud? (how many points?)
StuEv
I have made 50-60 posts
I have made 50-60 posts
Posts: 52
Joined: Mon Nov 16, 2015 12:52 am
8
Full Name: Julian
Company Details: University
Company Position Title: Student
Country: Germany
Linkedin Profile: No

Re: Deformation analysis statistics

Post by StuEv »

2 million points
Post Reply

Return to “Leica Cyclone 3DR”