- In order to scale a project, step 1. Initial Processing has to be completed.
- If the project has images with geolocation and the Scale Constraint is too different from the scale given by the image geolocation, the Scale Constraint will not be taken into account. If the scale given by the image geolocation is inaccurate, the geolocation should be removed to scale the project. For more information about the image geolocation: Menu Project > Image Properties Editor... > Image Geolocation.
In order to scale a project, a Scale Constraint need to be defined in the rayCloud:
1. On the Menu bar, click View > rayCloud.
2. On the Menu bar, click rayCloud > New Scale Constraint.
3. In the 3D View, a blue circle appears beside the mouse. Left click to insert the first vertex.
- A Scale Constraint is defined by a line connecting 2 vertices.
- For each vertex a Manual Tie Point is created.
- When drawing the Scale Constraint, a vertex used by other Objects (Polylines, MTPs) can be reused by clicking on it while drawing the Scale Constraint.
4. Right click to add the last vertex and to create the Scale Constraint.
5. To make the drawing more accurate, in the 3D View, select the Scale Constraint and correct the position of the vertices in the images: How to edit Objects in the rayCloud.
Note: For each vertex a Manual Tie Point is created. Each vertex that is marked on at least two images is taken into account in the Initial Processing if it is processed again or in the reconstruction if it is Reoptimized.
6. In the Selection section of the right side bar, enter the Initial Length.
7. (optional) Set the accuracy of the Initial Length in the box.
8. Click Apply to save the changes.
9. (optional) Add more Scale Constraints following steps 2. to 8.
10. Click Process > Reoptimize to scale the project.
Article feedback (for troubleshooting, post here instead)
Is more scale constraint is better ? For example, Orientation constraint need 3 max, so is 8 scales constraint is better dans 1 scale constraint ?
Having multiple-scale constraints is not recommended. The most important is finding a good edge/object to define the scale for better accuracy.
if i have to make a 3d model of a building without GCPs, how many times and in what positions is better to take misurements to scale the project and reach a good misure precision?
You can only have one scale constraint but make sure that the reference that you are marking is clearly visible and the marking is good. This should be enough.
I have a problem with scaling. After creating the scale bar, it should be reoptimized. I do it. But then the dense and sparse cloud disappear and the images are not aligned.
As you have already opened the ticket, it would be the best place to resolve this issue. Once the issue is solved, could you share your experience in this post so that it would be helpful for others?
Yes. Mr. Holden helped me a lot so that I could solve the problem. I did the steps they said and my problem was solved.
One of the important things to do is to adjust the size of the cameras and tie points (in Display Properties) at the end to fit the new scale.
Out of curiosity, how would I know when to scale a project? What should I look for? Is it best practice to always scale a project? And, do you have specific documentations on why one would need to do this theoretically?
Thanks so much!
Jenny Bloom If you do not have GCPs, the measurements will not be accurate. So scale constraints can be used to scale a project when GCPs are not an option. When a project is properly scaled that means you can be confident in the measurements that you take.