[GRASS-user] Re: grass-user Digest, Vol 25, Issue 61

[andrew haywood]

Hi Richard,

the lidar system emits a pulse of focused light in the near infrared region
of the electromagnetic spectrum and detects the radiation reflected back
from the target. Lidar intensity,  is the intensity of each return pulse
(ive seen it referred to as laser amplitude), it represents the reflected
energy and provides a concentrated measurements of the object's reflectance
unaffected by shadows or occlusions. This reflectance may vary based on the
reflectance properties  of the targeted material.

Most often in Australia the data suppliers provide a flatfile containing
x,y,x,i . where i is the intensity of the pulse. Each pulse is then
classified as ground or non ground (or vegetation in my case). Ive simply
taken the non-ground hits and removed the ground elevation (DTM) to get a
relative canopy height of my vegetation.

RE Stats package. I would like to use R but I have 1200 2kmby2km tiles with
approximately 3 billion points (approximately 700 Gb). r.in.xyz and its
statistics is an amazingly quick way to calculate summary statistics in a
grid format. Using a stats package just seems to have too much overhead. Ive
started to write a python program to read the flat files and calculate
summary statistics for the grid cells.

My problem is I can calculate 13 canopy height percentiles for my 20m
resolution grid cells really quickly with r.in.xyz- but then I wanted for
the same height percentiles the mean intensity values (basically a two
classification table on height percentiles and intensity) and was wondering
if it was possible to do this in grass easily.


cheers

Andy



On Thu, May 29, 2008 at 7:18 PM, Richard Chirgwin <rchirgwin at ozemail.com.au>
wrote:

>
>  ------------------------------
>>
>> Message: 7
>> Date: Thu, 29 May 2008 13:16:04 +1000
>> From: "andrew haywood" <ahaywood3 at gmail.com>
>> Subject: [GRASS-user] newbie: processing lidar with r.in.xyz
>> To: grass-user at lists.osgeo.org
>> Message-ID:
>>        <7a4569180805282016s3212c78ck2c35344b95650a7a at mail.gmail.com>
>> Content-Type: text/plain; charset="iso-8859-1"
>>
>> Dear List,
>>
>> i am currently enjoying the flexibility of using r.in.xyz to process lidar
>> data over native forests in Victoria, Australia.Thank you hamish for such
>> a
>> great tool.
>> At this stage I have 'binned' my data into 20m cells and calculated a
>> number
>> of metrics (including 13 vegetation height percentiles p5 p10 p20 .. p90,
>> p95, p99) based on the z values.
>> What I was wondering is how can I calculate the mean intensity for the
>> associated height percentiles.
>>
>> To get around this I have written a script in a proprietary stats package
>> to
>> create the intensity metrics. However, I would prefer to use Grass and
>> opensource tools to do this binning
>>
>> Any suggestions would be appreciated!!!!
>>
>> My data is in the following format
>>
>> x|y|ground|intensity|class|canopy_top|height
>> 410774.45|5820999.93|773.3|23|10|766.0983886719|7.20161
>> 410763.07|5820999.9|802.27|2|10|763.9434814453|38.3265
>> 410765.47|5820999.94|773.11|90|10|764.2877807617|8.82222
>> 410758.09|5820999.95|807.99|47|10|762.8372192383|45.1528
>> 410748.12|5820999.93|760.89|8|10|760.5731811523|0.316819
>>
>>
> Andrew,
> At the risk of sounding thick, I'd like to know what "intensity" represents
> in your context ... Regarding statistical packages, since R is grass-linked,
> why not do the calcs in R?
>
> Richard Chirgwin
>
>> ...
>> ...
>> ...
>>
>> I suspect I should start to learn python.
>>
>>
>> regards
>>
>> Andy
>> -------------- next part --------------
>> An HTML attachment was scrubbed...
>> URL:
>> http://lists.osgeo.org/pipermail/grass-user/attachments/20080529/5e7d669e/attachment-0001.html
>>
>> ------------------------------
>>
>> Message: 8
>> Date: Thu, 29 May 2008 11:17:48 +0800
>> From: "maning sambale" <emmanuel.sambale at gmail.com>
>> Subject: Re: [GRASS-user] Triangles polygons with Arc*
>> To: grass-user <grass-user at lists.osgeo.org>
>> Message-ID:
>>        <f902f9840805282017y2bae1830y786aa244a5dc60f6 at mail.gmail.com>
>> Content-Type: text/plain; charset=UTF-8
>>
>>
>>
>>> Just curious if anybody encountered this issue with Arc*:
>>>>
>>>>
>>> it's a common problem.
>>>
>>>
>> It does seem to be.
>>
>>
>>
>>> short answer: TINs stink. Raster maps created from TINs stink more.
>>>
>>>
>>
>> My colleague used SRTM DEM converted to grid format of Arc
>> He mentioned that it is fairly common when you convert raster grids in
>> arc to vectors (shapefiles).  Especially with very small areas like a
>> 2-3 pixels.   Will look into this further.
>>
>> To all the others who gave me a list of Arc mailing list thanks!  I am
>> reading a few right now.
>>
>> cheers,
>>
>> maning
>>
>>
>>
> _______________________________________________
> grass-user mailing list
> grass-user at lists.osgeo.org
> http://lists.osgeo.org/mailman/listinfo/grass-user
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.osgeo.org/pipermail/grass-user/attachments/20080530/512d4525/attachment-0001.html