Extract LAS Flight Line Tags

[BLMGIS]

Hi all,

Is it possible to derive a list of flight line tags from a merged las
file, and then create a new las file based on the value of flight line
tag?

For example, suppose the merged las file has 2 filght lines (line 1,
and line 2).

Is it possible to create a new las file from flight line 1, and
another new las file from line 2?

Thanks

[Martin Isenburg]

Hello Bureau of Land Management,

> Is it possible to derive a list of flight line tags from a merged las
> file, and then create a new las file based on the value of flight line
> tag?

yes. with the (undocumented) '-histo <attribute> <bin_size>' option of
lasinfo

lasinfo s1480690.las -histo point_source 1
[..]
point source id histogram with bin size 1
bin 36 has 336540
bin 37 has 108695
[..]

> Is it possible to create a new las file from flight line 1, and
> another new las file from line 2?

yes. with the '-keep_point_source <id>' filter of las2las

las2las -i s1480690.las -keep_point_source 36 -o s1480690_36.laz
las2las -i s1480690.las -keep_point_source 37 -o s1480690_37.laz

Regards,

Martin @lastools

[Martin Isenburg]

Hi,

while on the subject. I just got an inquiry for sample LAS/LAZ data
that are
1) a small number of overlapping flight paths
2) properly calibrated / aligned
3) classified for ground and non-ground pulses

you can download a lot of tiled flight paths in LAZ format that are
calibrated and classified from here

ftp://lidar.dnr.state.mn.us/data/raw/county/

you can them merge then on the fly with '-i *.laz' and get out the
flight-paths as follows:

lasinfo '-i *.laz' -merged -histo point_source 1

now look at what point_source IDs are of interest and extract them
with

las2las -i *.laz' -merged -keep_point_source 3001 -o fp3001.laz
las2las -i *.laz' -merged -keep_point_source 3002 -o fp3002.laz
las2las -i *.laz' -merged -keep_point_source 3003 -o fp3003.laz
las2las -i *.laz' -merged -keep_point_source 3004 -o fp3004.laz

and then sort the points along the flight path back into acquisition
order with

lassort -i fp30*.laz -olaz -gps_time

cheers,

martin @lastools

[Karl Heidemann]

Martin,

The timestamps are precise enough to be unique for each PULSE, but not for each POINT.

Will this final sort assure that the points' return number order for each "point family" also remain in ascending order?

If not, can you implement a 2-key sort, first on timestamp; second on RN ?

(yeah, yeah .... I know that the RN order SHOULD be OK to begin with .... BUT  ....)

Karl

H Karl Heidemann, GISP
Physical Scientist, Lidar Science
U.S. Geological Survey
Earth Resources Observation and Science Center
47914 252nd Street, Sioux Falls, SD 57198
Phn: 605-594-2861
Fax: 605-594-6529
kheid...@usgs.gov
http://lidar.cr.usgs.gov

 "Nothing matters very much, and few things ... matter at all."
                                                                -- Arthur James Balfour

 

 

From:	Martin Isenburg <martin....@gmail.com>
To:	LAStools - efficient tools for LIDAR processing <last...@googlegroups.com>
Date:	11/22/2011 04:47 PM
Subject:	[LAStools] Re: Extract LAS Flight Line Tags
Sent by:	last...@googlegroups.com

---

--
You can download LAStools at
http://www.cs.unc.edu/~isenburg/lastools/
You can visit the LAStools group at
http://groups.google.com/group/lastools/
You can be social with LAStools at
http://www.facebook.com/LAStools
http://www.twitter.com/LAStools

[Martin Isenburg]

Hello Yet Another Geospatial Government Agency,

(-;

> The timestamps are precise enough to be unique for each PULSE, but not for
> each POINT.

That is correct. Please be aware of that, folks!

> Will this final sort assure that the points' return number order for each
> "point family" also remain in ascending order?

Currently, for points that have the same GPS-time, the final sort will
maintain whatever order they had in the input. So if the points were
originally tiled with lastile.exe everything will be fine (except for
the rare case that first and second returns landed in different
files).

> If not, can you implement a 2-key sort, first on timestamp; second on RN ?

I could certainly add this. Maybe an additional '-return_number' flag
in the command line? I also noticed an inefficiency in my
instructions. You can directly extract & sort without generating an
extra copy on disk with

lassort -i *.laz' -merged -keep_point_source 3001 -extra_pass -
gps_time -o fp3001.laz
lassort -i *.laz' -merged -keep_point_source 3002 -extra_pass -
gps_time -o fp3002.laz
lassort -i *.laz' -merged -keep_point_source 3003 -extra_pass -
gps_time -o fp3003.laz
lassort -i *.laz' -merged -keep_point_source 3004 -extra_pass -
gps_time -o fp3004.laz

Regards,

Martin

[chuck]

Martin,

This link takes you to a list of counties. Within each county folder
there is a list of tiles that certainly include point records from a
set of overlapping flight lines. However, the number of flight lines
is not necessarily small nor are they necessarily all contained
within
that county.

So, I am still looking for

1) a small number of overlapping flight paths
2) properly calibrated / aligned
3) classified for ground and non-ground pulses

But really not looking for a clever way of backing out such a
data set.

Thanks for the suggestion.

Best Regards - Chuck O'Hara

[Crosby, Christopher]

We probably have data that meet the requirements. Try:
http://opentopography.org/id/OTLAS.032011.26912.1 or
http://opentopography.org/id/OTLAS.072011.26911.1 - both are relatively
narrow river floodplain surveys and should be composed of only a few
flight paths. Available as both LAS and LAZ.

Chris Crosby

http://www.opentopography.org/

[chuck]

Hi Chris,

Thanks so much for the reply and suggested data sources. I am now a
registered user at the OpenTopography site.

I went to the links, selected an area and found that if I want as many
as 150M pulses I need to be a power user so that request is likely in
your queue. It appears that my AOI request will likely generate a
single merged LAZ file for the area. If that is the case then -- I
will need to go back to Martin's suggested method of extracting by
source and then sorting. True? At the end - not sure I will end up
with a few good flightlines for rigorous testing.

Not whining or complaining - just really need some good candidate
data. Your help and suggestions certainly are welcome.

Regards - Chuck O'Hara

On Nov 22, 5:37 pm, "Crosby, Christopher" <ccro...@sdsc.edu> wrote:
> We probably have data that meet the requirements.  Try:http://opentopography.org/id/OTLAS.032011.26912.1orhttp://opentopography.org/id/OTLAS.072011.26911.1- both are relatively

> narrow river floodplain surveys and should be composed of only a few
> flight paths.  Available as both LAS and LAZ.
>
> Chris Crosby
>
> http://www.opentopography.org/
>

[Crosby, Christopher]

Yes, OT will return a merged LAS or LAZ file for your AOI and you will
need to extract the individual swaths via -keep_point_source as described
by Martin.

It sounds like what you want is the original swath files before they've
been merged and tiled? If so, send me a message off list and I can
provide a sample: ccr...@sdsc.edu

-cc