Martin,
Based on your announcement “Rapidlasso Announces LASzip “Compatibility Mode” For LAS 1.4”, it appears the laszip has supported a 1.4 to 1.2 “compatibility mode” for nearly three years. However, I don’t see an option for native 1.4 support in laszip.README.
The problems we are having with this are:
· We extract metadata reported by lasinfo to enter each LiDAR tile into a catalog database.
o The lasinfo report on a LAZ file created by laszip with a LAS1.4 input file has the following issues for us:
§ Reports a “version major.minor: 1.2”, instead of 1.4, and
§ the “*_extended” fields are missing, and
§ the histogram of classified points is missing
· Our customer contract requires us to deliver classified tiles in LAS 1.4 format, with a web delivery option. These files use the extended 256 classifications of point data record format 6.
o Naturally, a compressed format is preferable for web delivery. LAZ would be ideal, but the missing items from lasinfo (listed above) cause contractual issues for us. It would appear that we delivered the tiles in LAS 1.2 format and without classifications.
o Customer software designed to process the extended classifications in LAS 1.4 would not work. Is there a workaround for this?
o Our customers would therefore have to unzip the LAZ file to process the tile. This defeats the purpose of zipping as LAZ.
Is there a version of laszip that supports LAS 1.4 natively? If not, it appears that the only alternative is to use a neutral format such as zip, or gzip for our LAS 1.4 files. This would force the customer to unzip them prior to processing these files, thus, prevent support calls that would otherwise result from them trying to process these LAS 1.4 files in “compatibility mode” LAZ format.
I created the LAZ file from a LAS 1.4 file (16206628.las) as follows:
laszip -i 16206628.las -o 16206628.laz
Attached are the lasinfo outputs of both files.
Tom Jordan, MSCS
Software EngineerGeospatial data processing and visualization
Space and intelligence systems / HARRIS CORPORATION
Melbourne, FL / HTC-C3014
Martin,
Thank you for your rescale recommendation of:
laszip -i 16206628.las -rescale 0.01 0.01 0.01 -auto_reoffset
That reduced the LAZ file size to just 64% of LAZ made without the rescaling option. And yes, 1 cm (0.4 inch) precision should be more than sufficient. Good catch!
However, the “LASzip compression” is not what you expected. Lasinfo reports this:
LASzip compression (version 3.0r3 c2 50000): POINT10 2 GPSTIME11 2 BYTE 2
> Martin Isenburg wrote:
> Note that how if you compress your LAS 1.4 files with the latest laszip.exe
> coder this line in your lasinfo report
> LASzip compression (version 3.0r3 c2 50000): POINT10 2 GPSTIME11 2 BYTE 2
>
> will change to
> LASzip compression (version 3.1r0 c3 50000): POINT14 3
I believe I am using the latest version of LAStools since the first line in the CHANGES.txt file is:
“26 July 2017 -- las2tin: now supports DXF format as output option for the generated Delaunay TIN”
(I see that ‘LASzip DLL: “native LAS 1.4 extension”‘ was introduced 25 April 2017.)
Also I’m still seeing these issues with the lasinfo output on this LAZ file generated with laszip:
1. “version major.minor: 1.2”, instead of 1.4
2. the “*_extended” fields are missing
a. lasinfo correctly reports this for the original LAS1.4 file:
start of waveform data packet record: 0
start of first extended variable length record: 0
number of extended_variable length records: 0
extended number of point records: 78585827
extended number of points by return: 78585827 0 0 0 0 0 0 0 0 0 0 0 0 0 0
...
extended_return_number 1 1
extended_number_of_returns 1 1
extended_classification 1 81
extended_scan_angle 2500 2500
extended_scanner_channel 0 0
b. Those sections are missing from the lasinfo of the LAZ file.
3. the histogram of classified points is missing
a. lasinfo correctly reports this for the original LAS1.4 file:
histogram of extended classification of points:
112 extended classification (70)
733 extended classification (72)
757 extended classification (80)
32 extended classification (81)
b. That section is missing from the lasinfo of the LAZ file.
Both lasinfo outputs are attached. Am I doing something wrong?
Thanks