Optimize Scans After a Registration in SCENE
Overview
This article covers the scan optimization procedure after a registration. You can only run this compensation after registering clusters with cloud-to-cloud connections between scans. You cannot run this compensation if you have used only a target-based registration.
What is Compensated by this Process
We analyze the geometry within the overlap areas of multiple scans to compensate deviations in the scans. Such deviations may have different root causes including a not properly warmed up scanner, or dirty optics. See the user manual of the laser scanner for more details.
What is not Compensated by this Process
This process does not compensate the angular errors in the scanner. If you detect a gap, for example double walls in the overlap area between the beginning and the end of the scan, you must run an on-site compensation on your scanner. For more information, see your scanner user manual.
How to Identify Scans that are Affected by these problems
After you have run a registration, you will notice some gaps, for example double walls, between the registered scans during the visual verification, see also figure below. You cannot fix these gaps as usual, i.e. by changing the registration parameters or by adding or removing connections between the scans.
Run the Scan Optimization
Before you start the procedure, make sure that the scans are not affected by the angular error described above.
- Go to the interactive registration.
- Open the registration view of the cluster.
- Unlock the Scan Manager of the cluster for which you want to run the post-registration optimization, including the subordinate clusters.
- Right-click the cluster in the structure view and select Operations > Registration > Post Registration Scans Optimization.
- The compensation process will start. When the process is finished, a message will be displayed telling you how much the registration mean error has been improved.
- If no scans were found for optimization, the message will be: No scans found that can be optimized.
- If large errors were found, the algorithm will be able to improve the scans. Despite these improvements, the scans' accuracy will still be compromised. Such errors occur, for example, when a scanner mirror is extremely dirty. In such a case, the following error message will be displayed:
- If the scans in the cluster need to be compensated, the process will take the equivalent time to process and register (refine connections and update statistics) the scans from the beginning.
The results of the compensation process are:
- Compensated scans
- The registration is updated using the new compensation scans