Hi All, I am searching for a way to extract targets from a point cloud. I would like a similar mathematical fitting of the targets to the points by selecting a section of points from the overall cloud, not by selecting the points from the station image as is done in the traditional target based registration.
Does anyone know of a way to do this? I am using version 11.
Targets from Points
-
- I have made <0 posts
- Posts: 7
- Joined: Fri Mar 04, 2022 6:57 am
- 2
- Full Name: Roger Moore
- Company Details: Land Registry Services
- Company Position Title: Junior
- Country: Australia
- Linkedin Profile: No
- paranoiahasu
- I have made 40-50 posts
- Posts: 44
- Joined: Wed Jul 16, 2014 9:02 am
- 9
- Full Name: ANIL
- Company Details: Mergen3D
- Company Position Title: Managing Director
- Country: Germany
- Linkedin Profile: Yes
- Has thanked: 50 times
- Been thanked: 6 times
- Contact:
Re: Targets from Points
Have you tried the Auto Target Extraction option in the Target-Based Registration menu?
-
- I have made 60-70 posts
- Posts: 60
- Joined: Wed Jun 12, 2013 10:59 am
- 10
- Full Name: Tony Millington
- Company Details: ISIS Rutherford Appleton Laboratory
- Company Position Title: Survey section leader
- Country: uk
- Linkedin Profile: Yes
Re: Targets from Points
I can think of two ways to do this, both of which probably only work with sphere targets - but I'm not sure what you gain by doing so.
1 - Run the Target based registration as usual, and then work through selecting the individual point clouds TRW creates for each target and merge those that are for the same target, you can then fit a sphere to the merged point cloud.
2 - Register the project and then use the 'Cloud based Modeler' option with the registered full point cloud to select the points you want each time and fit a sphere.
I guess the aim is to use all the available points to fit a target, rather than just those from one station scan, to try and better model the 'correct' position for a sphere?
Whether this gives a 'better' result than the final mean target positions from the Target based Registration, I'd be interested to know.
1 - Run the Target based registration as usual, and then work through selecting the individual point clouds TRW creates for each target and merge those that are for the same target, you can then fit a sphere to the merged point cloud.
2 - Register the project and then use the 'Cloud based Modeler' option with the registered full point cloud to select the points you want each time and fit a sphere.
I guess the aim is to use all the available points to fit a target, rather than just those from one station scan, to try and better model the 'correct' position for a sphere?
Whether this gives a 'better' result than the final mean target positions from the Target based Registration, I'd be interested to know.
-
- I have made <0 posts
- Posts: 7
- Joined: Fri Mar 04, 2022 6:57 am
- 2
- Full Name: Roger Moore
- Company Details: Land Registry Services
- Company Position Title: Junior
- Country: Australia
- Linkedin Profile: No
Re: Targets from Points
It's an imported e57 file based on a mobile scan. I'm wanting to compare the positions of multiple scans of the same room to measure its accuracy.
Auto Target Extraction doesn't find any targets.
The cloud based modeller idea indeed worked for the spheres I had placed which gives me something to work with. Unfortunatly I only located the paper targets with a Total Station so I'm still searching for that solution.
Auto Target Extraction doesn't find any targets.
The cloud based modeller idea indeed worked for the spheres I had placed which gives me something to work with. Unfortunatly I only located the paper targets with a Total Station so I'm still searching for that solution.