LAStools returning a whole number for average Intensity values

[Dwight Crouse]

Hi Martin,

Just wondering why LAstools returns a whole number for average intensity values?  Shouldn't it have some decimals?  I'm using version 170102.

Cheers,

Dwight

[Dwight Crouse]

Just to clarify, I'm using a lascanopy in the following syntax:

lascanopy -v -i *.laz -o first_ht_ge_1m.csv -keep_first -files_are_plots -names -height_cutoff 1.0 -avg -std -int_avg -int_std 

Thanks,

Dwight

Dwight Crouse
SENIOR DATA ANALYST

Tel: 403.836.9429 | Tollfree: 1.866.698.8789 ext. 4
www.tesera.com

Google Plus | LinkedIn

--
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

[Martin Isenburg]

Hello Dwight,

It seemed that given the main use of intensity rasters or averages and the questions surrounding the exact meaning of these intensity values a resolution of rounded to the nearest integer would be sufficient. Do you have an example use case where a higher resolution average is needed? I would be curious to know about it.

http://rapidlasso.com/lascanopy

http://lastools.org/download/lascanopy_README.txt

Regards,

Martin

[Dwight Crouse]

Hello Martin,

We're investigating different techniques to use intensity values to get the best results for feature extraction, species and landcover classifications and we think having this extra resolution on intensity could help.  I'm using LAStools but also FME to develop additional calculations with the indices derived from LAStools, so it would help to have the actual averages prior to rounding since I'm doing some transformations on the data.  Maybe you could add a new average function to the lascanopy tool to retain the mathematical average for those that might need it?  We're also using additional indices from LiDAR and Colour Infrared Imagery to assist with feature extraction, species & landcover classifications as well.  

Regards,

Dwight

[Perdue, Michael]

Hello,

 

You can get what you want from lastools in a couple of indirect ways. Probably the easiest is to pass the file through a text file and swap "z" for "i".

 

Using fuza.laz as an example;

:: Convert fusa.laz from elevations to heights.

lasheight -replace_z ^

          -i fusa.laz ^

          -o fusa_1.laz

 

:: Extract xyi from all points in fusa_1.laz. Here, I simulate a breast height of

:: 1.37m (-drop_z_below 1.37) to be in alignment with lascanopy’s defaults.

las2txt -parse xyi ^

-drop_z_below 1.37 ^

        -i fusa_1.laz ^

        -o fusa_1.txt

 

:: Use lasgrid instead of lascanopy to calculate the metrics. We use

:: -elevation_average instead of -intensity_average and spoof lasgrid by telling

:: it the third column is elevation instead of intensity. This way the value is

:: not rounded to the nearest integer and stored in a 8/16bit container. To align

:: with lascanopy I use a step size of 20m, but choose as you see fit.

lasgrid -step 20 ^

        -elevation_average ^

        -i fusa_1.txt ^

       -iparse xyz ^

        -o fusa_intensityAverage.tif

 

Cheers,

 

Mike

 

From: last...@googlegroups.com [mailto:last...@googlegroups.com] On Behalf Of Dwight Crouse
Sent: Tuesday, January 24, 2017 8:29 AM
To: LAStools - efficient tools for LiDAR processing <last...@googlegroups.com>
Subject: Re: [LAStools] LAStools returning a whole number for average Intensity values

 

Hello Martin,

 

We're investigating different techniques to use intensity values to get the best results for feature extraction, species and landcover classifications and we think having this extra resolution on intensity could help.  I'm using LAStools but also FME to develop additional calculations with the indices derived from LAStools, so it would help to have the actual averages prior to rounding since I'm doing some transformations on the data.  Maybe you could add a new average function to the lascanopy tool to retain the mathematical average for those that might need it?  We're also using additional indices from LiDAR and Colour Infrared Imagery to assist with feature extraction, species & landcover classifications as well.  

 

Regards,

 

Dwight

On Wednesday, January 18, 2017 at 7:12:50 PM UTC-7, Martin Isenburg wrote:

Hello Dwight,

 

It seemed that given the main use of intensity rasters or averages and the questions surrounding the exact meaning of these intensity values a resolution of rounded to the nearest integer would be sufficient. Do you have an example use case where a higher resolution average is needed? I would be curious to know about it.

 

http://rapidlasso.com/lascanopy

http://lastools.org/download/lascanopy_README.txt

 

Regards,

 

Martin

On Jan 19, 2017 6:57 AM, "Dwight Crouse" <dwight...@tesera.com> wrote:

Just to clarify, I'm using a lascanopy in the following syntax:

 

lascanopy -v -i *.laz -o first_ht_ge_1m.csv -keep_first -files_are_plots -names -height_cutoff 1.0 -avg -std -int_avg -int_std 

 

Thanks,

 

Dwight

 

 

[Dwight Crouse]

Hi Martin and the rest of the group,

The process that Michael Perdue provided worked for creating an average intensity for the plot with the resolution that I wanted but I also need to create intensities at different deciles.  Looking at the gridcell readme file I'm not sure if it is possible to summarize data in lasgrid using deciles?  It would be some programming to do what I want in python, so I'm wondering if I'm back to requesting some extra resolution in the various intensity outputs within lascanopy, any thoughts out there in the group?  I'd like to keep my process concentrated in LAStools if at all possible just for future maintenance concerns.  

Dwight

--

Dwight Crouse
SENIOR DATA ANALYST

Tel: 403.836.9429 | Tollfree: 1.866.698.8789 ext. 4
www.tesera.com

Google Plus | LinkedIn

[Martin Isenburg]

Hello,

to more easily realize intensity averages with more decimal digits we'll add a LAStransform to the next release of LAStools that will allow to copy the intensity into the z coordinate on-the-fly with '-copy_intensity_into_z'. Then you would be able to get all your intensities with higher precision as averages and percentiles using lascanopy. Of course ... getting the same for the elevation would require a second run.

http://rapidlasso.com/lascanopy

http://lastools.org/download/lascanopy_README.txt

But adding minor modifications to existing metrics in lascanopy is not something I am willing to do lightly if there are needed by very few (just you?) and there are easy work-arounds like described above.

​

How are intensities working for you? Are you normalizing them somehow in order not to run into such issues:

https://www.facebook.com/LAStools/photos/846786945383751

https://www.facebook.com/LAStools/photos/846786915383754

https://www.facebook.com/LAStools/photos/846786895383756
https://www.facebook.com/LAStools/photos/846786865383759

Do you have a way to detect cloud or smog that both can modify the intensity values significantly?

https://www.facebook.com/LAStools/photos/849985711730541

Regards,

Martin @rapidlasso

[Sam Hackett]

Hi Dwight,

Just wondering if you started by scaling up intensities in the source data you might avoid a need for decimal places later on?

Sam