Split LAS by order and location

[JaeO]

Hello All,

I have LAS files derived from single photon data that lack any of the common attributes outside of x,y,z; no scan angle, flight line, return number, etc. Would anyone have a suggestion on how I might tease out flight line information? It seems that the data was written to the LAS file in the order in which it was collected, so I am wondering if it would be possible to scan through the file by order written and when the distance between two consecutive points is greater than some threshold, flag as new flight line. Does anyone know if this is possible within the LAStools suite? I would really rather not write my own code to do this if it exists elsewhere.

[Martin Isenburg]

Hi Jae,

it would be reasonably easy to add an alternate '-recover_flightline' option to lassplit that does not work with GPS time but with jumps in the x/t distance of the points. Can you affirm that those flightlines are really in acquisition order and that flight line jumps really exist and that they are big enough compared to the regular x/y distances? What is the scan pattern you see? Is is data from the Sigma Space system? A few pictures (> 500 KB) would be nice that show the scan pattern and the "jump".

You can output the difference in subsequent x and y and z coordinates with las2txt using:

las2txt -i spl.laz -parse )!@

where )!@ represent the delta of the xyz coordinate to the previous point:

E:\LAStools\bin>las2txt -i ..\data\fusa.laz -parse )!@ -stdout | more

27799997 612234220 6435

0 33 -5

-45 -279 -121

3 31 -21

-2 36 12

0 34 4

2 32 -18

-7 74 60

37 58 -339

-21 91 186

-3 38 29

-4 39 37

-7 41 65

0 34 -2

2 30 -21

0 34 6

1 33 -14

-1 36 14

2 63 -24

7 26 -60

1 33 -17

-2 35 25

1 33 -9

[...]

On Thu, Jan 12, 2017 at 5:48 PM, JaeO <jae.o...@gmail.com> wrote:

Hello All,

I have LAS files derived from single photon data that lack any of the common attributes outside of x,y,z; no scan angle, flight line, return number, etc. Would anyone have a suggestion on how I might tease out flight line information? It seems that the data was written to the LAS file in the order in which it was collected, so I am wondering if it would be possible to scan through the file by order written and when the distance between two consecutive points is greater than some threshold, flag as new flight line. Does anyone know if this is possible within the LAStools suite? I would really rather not write my own code to do this if it exists elsewhere.

--
Download LAStools at
http://lastools.org
http://rapidlasso.com
Be social with LAStools at
http://facebook.com/LAStools
http://twitter.com/LAStools
http://linkedin.com/groups/LAStools-4408378
Manage your settings at
http://groups.google.com/group/lastools/subscribe

[Olsen, Richard (RC) (CIV)]

Jae,

  I suspect the answer is “no” for reasons that will not be obvious in the data.   At least for the Sigma Space sensor, the data really should be divided up into forward and aft scans, and probably also by some of the instrument internals (how they store the data).   You are going to need more information – and in particular time if you can get it.  The SS sensor uses a scan mirror, and you need the scan information to properly interpret the data.

 

RC Olsen

Naval Postgraduate School

--

[Terje Mathisen]

What do you need it for, and how accurate does it need to be?

Assuming aquisition order (like Martin writes) is crucial, but with that
given you could easily determine flight paths, flight line limits,
scanning mirror type (back & forth vs rotating mirror).

With the midpoint of each flight line as the flight path you can then
guess which pulses have multiple returns, since they should all be on a
fairly straight line in (x,y,z) space, adn then you can use those
multi-return pulses to calculate the flight height. :-)

This sounds like a fun programming challenge, I can start with a normal
LAZ file sorted in aquisition and dump just xyz to a text file, then see
how well the missing info can be recreated.

Terje

--
- <Terje.M...@tmsw.no>
"almost all programming can be viewed as an exercise in caching"

[Stoker, Jason]

Terje, It may not be impossible (especially with some assumptions), but it be a bit more complicated than you think- the Sigma instrument as I understand it (HRQLS) has a 10 × 10 array of 100 beamlets, generated by a passive diffractive optical element (DOE) in front of a 25 kHz laser. It is a combination of a conical scanning mirror with these beamlets. The returned 100 beamlets are imaged into 10 × 10 micro-channel plate PMDs, which means you theoretically could get an array of 100 returns all at the same time (in different XYZ).  There are not any multiple returns. As RC Olsen said, it is a bit of a different animal than traditional linear-mode sensors. They may even use a combination of fore and aft looks to determine signal (points) from noise.

Some of the system specs of HRQLS can be found here in this paper: http://www.mdpi.com/2072-4292/8/9/767/htm

I would be interested in any way to tease this information out, so please keep me in the loop!

Jason M. Stoker, Ph.D
US Geological Survey
National Geospatial Program
Office: 970-226-9227
Cell: 605-496-3513
My USGS Profile 

[Terje Mathisen]

Stoker, Jason wrote:
> Terje, It may not be impossible (especially with some assumptions),
> but it be a bit more complicated than you think- the Sigma instrument

> as I understand it (HRQLS) has a 10 × 10 array of 100 beamlets,

> generated by a passive diffractive optical element (DOE) in front of a
> 25 kHz laser. It is a combination of a conical scanning mirror with

> these beamlets. The returned 100 beamlets are imaged into 10 × 10

> micro-channel plate PMDs, which means you theoretically could get an

> array of 100 returns all at the same time (in different XYZ). There

> are not any multiple returns. As RC Olsen said, it is a bit of a
> different animal than traditional linear-mode sensors. They may even
> use a combination of fore and aft looks to determine signal (points)
> from noise.

Ouch! This is a _lot_ more complicated than I guessed. Good thing I
didn't ASSuMe and just started programming.
>
> Some of the system specs of HRQLS can be found here in this paper:Â
> http://www.mdpi.com/2072-4292/8/9/767/htm

Interesting animal...

>
> I would be interested in any way to tease this information out, so
> please keep me in the loop!

Well, not from this particular instrument I'm afraid. :-(

Terje

>
> Jason M. Stoker, Ph.D
> US Geological Survey
> National Geospatial Program
> Office: 970-226-9227
> Cell: 605-496-3513

> My USGS Profile <https://www.usgs.gov/staff-profiles/jason-stoker>

> - <Terje.M...@tmsw.no <mailto:Terje.M...@tmsw.no>>

> "almost all programming can be viewed as an exercise in caching"
>
> --
> Download LAStools at
> http://lastools.org
> http://rapidlasso.com
> Be social with LAStools at
> http://facebook.com/LAStools
> http://twitter.com/LAStools
> http://linkedin.com/groups/LAStools-4408378
> <http://linkedin.com/groups/LAStools-4408378>
> Manage your settings at
> http://groups.google.com/group/lastools/subscribe

> <http://groups.google.com/group/lastools/subscribe>

[JaeO]

Hello All, The reason for this request is in an attempt to resolve some ranging issues associated with the dataset. See my earlier post for information on the problem. If any of you bright minds have any suggestions other than manual point classification, I would be very excited to hear them.

[Martin Isenburg]

Hello Jae,

nevertheless the "geometric jump" way of segmenting out the flight lines may still work ... we may just have to look for bigger jumps. This would be most robust if your tiles (?!?) are very big (10 km by 10km?) and in acquisition order so that those beams of 100 beamlets enters and exists the tile at very distant locations (withing that large tile). What I do not understand is why no flight line index was provided by the vendor? It seems that you got the data in LAS and while there are many fields in LAS that do not have a corresponding meaning in SPL data ... flight lines (aka the population of the point source ID of each point) does! Who provided this data to you? Can you provide us with a link to a sample as you obtained it from your source?

Martin @rapidlasso