when i try to import a E57 laser scan it looks fragmented. Even if i try to reconstruct. First i thought the viewport is limited but it seems like points far away from the scan position are cut off. Look at the ceiling parts. The Scan is from Z+F registered in cyclone and exported as E57. The Scan has about 177million points and 4GB in file size. Different import settings did not change anything. The E57, lsp, rcproj file is online and ready for sharing via dropbox.
Any idea why its not working? Is it a E57 export Problem? I think cyclone has no options for exporting E57 in different ways.
RC limits the visible points in the 3D viewport. Raise the value (carefully) in Workflow - Application - settings - Max points to display. It should recalculate and then you should see more points.
If you click on the component, what does it say at Points count?
Thank you for your reply but increasing the max point setting does not help. In my settings i set max points to display from 100.000.000 to 500.000.000 million. The amount of points displayed in the viewport does not change.
If i click on the component the point count is 240.000
I have a theory. The 240.000 are the clue. From the screenshots I gather it’s only one station, right?
If so, then it would confirm my theory that RC converts each laserscan into 6 images represented as a little cube at the scanner location. And 240.000 divided by 6 is 40.000 which is the standard max features per image setting. So if you raise that to say 30.000.000 (alignment settings), you should get all the features (if I’m right) since 30 mil x 6 = 180 mil. Maybe you also need to raise the mpx setting. It might be neccessary to delete all components prior to a new alignment before it works.
If you then add images into the mix, it will be difficult because the setting means that basically all detected features in one image will be used, also the very very bad ones.
I am sure that if you add more scans, it would work better even with a lower setting - I think 80.000 is a recommended maximum.
@Götz Echtenacher thanks again for your reply. You are right RC shows 4 image cams in the middle of the laser scan location.
I increased both mpx and max feature per image but also with very high values i get never more than 80.000 features per image.
. I
Also tested more than one scan (100 scans from indoor) but always the ceiling for example is fragmented and i know there million points which are very good in precision.
I dont know if i am on the right path to mesh these laser scans in RC i thought it can handle that.
How about you just reconstruct a small portion of the main vaults and then you should see what’s going on there. If the geometry is there, the mesh should be ok. If you plan to add images, I would do that for a small area before reconstruction. Same here, if the geometry is there, RC should align them in the correct location.