Re: [LAStools] inquiry of clipping of LAS 1.3 FWF data

41 views
Skip to first unread message

Martin Isenburg

unread,
Mar 25, 2019, 5:16:40 PM3/25/19
to LAStools - efficient command line tools for LIDAR processing, PulseWaves - no pulse left behind
Hello,

so Genping shared a pair of LAS/WDP files with me. The mission is to clip out a LAS slice as well as its corresponding WDP part from these huge (2GB LAS, 8GB WDP) files . First I ran lasinfo (see attached report).

2,645,937,193 NP_T-0892_FWF.las
8,133,849,280 NP_T-0892_FWF.wdp

lasinfo -i NP_T-0892_FWF.las -odix _info -otxt

The easiest way to get only a small part of LAS plus WDP from this file is to use laszip with the special option '-waveform_with_map' in combination with the filters that are part of every of the LAStools. For example the coordinate filters, GPS time filters, scan angle filters or any combination of them.

[...]
  -keep_tile 631000 4834000 1000 (ll_x ll_y size)
  -keep_circle 630250.00 4834750.00 100 (x y radius)
  -keep_xy 630000 4834000 631000 4836000 (min_x min_y max_x max_y)
[...]
  -keep_scan_angle -15 15
[...]
 -keep_gps_time 11.125 130.725
[...]

First I use GPS time filter to slice out 5 milliseconds worth of LiDAR using some time stamp range that falls into the GPS time range reported in the lasinfo file:

laszip -i NP_T-0892_FWF.las ^
          -keep_gps_time -999516650.000 -999516649.995 ^
          -waveform_with_map ^
          -o NP_T-0892_FWF_483350_005ms.las

This extracts these two files and a visual check with lasview gives the desired result that are also attached. However, Because laszip does not change the bounding box we first need to run lasinfo to tighten the bounding box to enclose only the few returns of the file. In the first of the three images I zoom out and draw "pulses" through all multi-returns that go 1000 meter in direction of the plane. We can nicely see where the plane must have been flying.

  76,534 NP_T-0892_FWF_483350_005ms.las
219,026 NP_T-0892_FWF_483350_005ms.wdp

lasinfo -i NP_T-0892_FWF_483350_005ms.las -repair_bb

lasview -i NP_T-0892_FWF_483350_005ms.las 

================================================================

Instead of producing LAS and WDP we can also produce LAZ and WDZ files which are much smaller.

laszip -i NP_T-0892_FWF.las ^
          -keep_gps_time -999516650.000 -999516649.995 ^
          -waveform_with_map ^
          -o NP_T-0892_FWF_483350_005ms.laz 

  22,599 NP_T-0892_FWF_483350_005ms.laz
  60,723 NP_T-0892_FWF_483350_005ms.wdz

lasinfo -i NP_T-0892_FWF_483350_005ms.laz -repair_bb

lasview -i NP_T-0892_FWF_483350_005ms.laz

So small - in fact - that we can attach these 5 milliseconds worth of LiDAR to this email. To produce the visualizations shown you need to hover over a point with the mouse cursor and press 'i' to select it. Then press SHIFT-'w' (or 'W') to activate the display of full waveforms. You will need to zoom in to see them. Use the pop-up menu to enable drawing pulses of 200, 500, or 1000 meters.

Regards from Costa Rica.

Martin

On Fri, Mar 22, 2019 at 2:35 AM Martin Isenburg <martin....@gmail.com> wrote:
Hello,

it is possible but requires some trickery. Can you share a sample file that I could look into via Dropbox or GoogleShare or WeTransfer? The WARNING are typical for export software that populates the bounding box with double-precision float values that are not quantized to the scalar factors (which are millimeter in your case). However ... it seems (some of) your wavepacket direction vectors are broken ... when I look at the min max summary I see some very wrong values (copied below):

  Wavepacket Index    1 3
             Offset   0 8466976000
             Size     160 480
             Location 1566.65 228567
             Xt       -1.#INF 1.#INF
             Yt       -1.#INF 1.#INF
             Zt       -3.87078e+031 2.7303e+029

That is of real concern. Are you able to visualize the waveforms with lasview of LAStools or pulseview of PULSEtools?

Regards from Costa Rica,

Martin @rapidlasso

On Fri, Mar 22, 2019 at 1:55 AM Genping Zhao <gen...@ualberta.ca> wrote:
Hi there,

I am a very new learner on LiDAR data processing. I got a LAS 1.3 FWF data, and would like to clip it to get the region of interested as another hyperspectral data. I attached the las file information in attachment. So does anyone can tell me how to or what free software can be used to do the clipping and registration? I used software LiDAR360 to clip it, but the derived clipped data without any waveform information.  

By the way,I am wondering if the warnings obtained by lasinfo  mattering a lot?Looking forward to your kind help and many thanks!

Best regards,
Genping

--
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
NP_T-0892_FWF_info.txt
NP_T-0892_FWF_483350_005ms_full_waveform_00_pulses_1000m.png
NP_T-0892_FWF_483350_005ms_full_waveform_01.png
NP_T-0892_FWF_483350_005ms_full_waveform_02.png
NP_T-0892_FWF_483350_005ms.laz
NP_T-0892_FWF_483350_005ms.wdz

Martin Isenburg

unread,
Mar 26, 2019, 11:29:13 AM3/26/19
to LAStools - efficient command line tools for LIDAR processing, PulseWaves - no pulse left behind
Hello,

here is another example using '-keep_circle' to cut out a circle with 12 meter radius (24 meter diameter) from somewhere in the middle of the flight line:

laszip -i NP_T-0892_FWF.las ^
          -keep_circle 794180 9782200 12 ^
          -waveform_with_map ^
          -o NP_T-0892_FWF_794180_9782200_12m.laz
WARNING: number of samples for wave packet descr 4 is zero
WARNING: number of samples for wave packet descr 5 is zero
WARNING: number of samples for wave packet descr 6 is zero
WARNING: number of samples for wave packet descr 7 is zero
WARNING: number of samples for wave packet descr 8 is zero
WARNING: number of samples for wave packet descr 9 is zero
WARNING: number of samples for wave packet descr 10 is zero
WARNING: written 4205 points but expected 46419930 points

lasinfo -i NP_T-0892_FWF_794180_9782200_12m.laz -repair_bb

lasview -i NP_T-0892_FWF_794180_9782200_12m.laz

The results are attached as pictures and data.

Regards,

Martin @rapidlasso
NP_T-0892_FWF_794180_9782200_12m_full_waveform_01.png
NP_T-0892_FWF_794180_9782200_12m_full_waveform_02.png
NP_T-0892_FWF_794180_9782200_12m_full_waveform_03.png
NP_T-0892_FWF_794180_9782200_12m.laz
NP_T-0892_FWF_794180_9782200_12m.wdz
Reply all
Reply to author
Forward
0 new messages