Low density data is often the name of the game with laser scan surveys and as SUA cover and simultaneous location and measurement (SLAM) systems emerge the dreaded dot to dot drawing game gets unavoidable.
ZebRevo will soon have serious competition from Leica’s BLKGO device. The SLAM age is dawning and that’s good news for anyone who has to solve internal measurement of buildings.
The SLAM data is very ‘thin’ despite spatial precision the clarity is a function of the height or thickness of the slice: the thicker it is the more you see a ‘line’
A look at this slice of the window shows the nature of the problem: the data is just too thin to work up into a set of meaningful profiles.
A measured drawing makes things clear:
Plotting the profiles from the measured drawing into the overall geometery of the point cloud resolves the detail:
SLAM data seems to be consistently ‘light’ on edge definition, ‘classic’ laser scans done with appropriate control and density are a boon, I can extract moulding profiles with relative ease:
but at high level the story changes, a first floor window scanned from ground level starts to get difficult: the upper surfaces of the cills are lost and the mouldings are lost on the ‘blind’ side of the scanner:
Developing an elevation drawing for 1:50 scale is possible but meeting a 1:20 scale resolution is unlikely:
Compared to working with point clouds an ortho-image is a much more accessible data form. Interior elevations from SLAM lack detail but infill photography can make all the difference:
Even mangled by SFM the photography has perspective resolved in the ortho-image and digitising is straight forward:
Selection of line and line weight are of course subjective and in this case driven by a 1:50 scale presentation.
The metric performance of the point clouds provided by SLAM have great potential provided an appropriate response to detail is in place. The camera and the measured sketch are as crucial as ever.