possible issues or lasboundary limitations?

[chuck]

Hi Martin,

When processing a very large LAS file through lasboundary, with holes
enabled and a small value for concavity, if there is a problem does
the tool currently attempt to write an 'errorlog.txt' file in some
location? This may not be your software, but I am getting these
messages thrown in this or a related process. Is this part of the
utility or extraneous?

Otherwise, have you run into any limitations or do you have any
guidance on the upper size of LAS the file, the maximum number of
pulses, the smallest value for concavity, or the maximum number of
polygons (if holes are on) or other aspects or limitations of
lasboundary that you can share with users?

Thanks for your help!

Chuck

[Martin Isenburg]

Hello,

after exchanging an off-list email with Chuck he told me he was using
lasboundary with setting '-concavity 5' and '-holes'. This means that
lasboundary will try to generate a boundary polygon with a granularity
of edge lengths as small as 5 units. The data set he was trying this
on was in scaled in feet where the default concavity would have been
150 (the default for meters or unspecified units is 50).

Running lasinfo.exe quickly revealed that his setting '-concavity 5'
was way too small. Here an excerpt from running lasinfo with the
option '-compute_density' ...

[...]
number of point records: 66,705,904
[...]
min x y z: 12991192.425 588397.501 611.122
max x y z: 13142242.349 594146.283 3032.417
[...]
key 2052 GeogLinearUnitsGeoKey: Linear_Foot
[...]
number of last returns: 58,426,231
covered area in square feet/miles: 768,445,344/27.56
point density: all returns 0.09 last only 0.08 (per square foot)
point spacing: all returns 3.39 last only 3.63 (in feet)
[...]

Since there is only one LiDAR point per 0.09 square foot or - as seen
the other way around - the point spacing is only 3.39 feet, a
concavity of 5 feet will create an insanely "fractal" polygon (that
grows deep channels into the data set and contains almost all points)
which will consume enormous amounts of memory and cause lasboundary to
run out of memory. As a rule of the thumb your '-concavity xxx' value
should be 10 - 50 times higher than the point spacing.

Martin @lastools

PS: Later today's release 120222 will have the lasinfo.exe version
that reports the point spacing. the current release only reports the
point density.