[GRASS-user] Stereoscopic images
Amagine
amagine at telus.net
Sun Jul 1 04:04:56 EDT 2007
I have not had enough time to seriously spend finding an OpenSource
alternative.
UNBC GIS Lab (University of British Columbia) http://
www.gis.unbc.ca/ They are doing some fantastic work there.
uhmm I believe Stacy Porter did some work which is similar in nature
to what you are wondering Morten.
http://www.gis.unbc.ca/courses/geog499/projects/2005/proters1/index.html
yes, and it appears she used PCI Geomatics orthoengine for extraction.
Here are a couple links to investigate further:
http://www.pcigeomatics.com/cgi-bin/pcihlp/OESE
http://www.pcigeomatics.com/cgi-bin/pcihlp/PACE
Now, there is a very old opensource solution that can also do the
parallax maths I believe, although I have not been able to test it,
and I do not know anyone who has.
I am sure the theory is solid though.
You can also check out http://mpa.itc.it/markus/spectral_unmixing/
about spectral angle mapping.
Thanks to Markus Neteler and company from Trento Italy,
not too sure how that project developed but seeing as he replied
here ...
Anyways there is a good paper in the above link covering image
registry using i.ortho.photo (?) in GRASS.
And you may also want to check out OpenDX, they may have some tricks
over there as well.
http://www.opendx.org
There is some really good info on methodology of photogrametry
analysis there anyways: ie. http://activetectonics.asu.edu/LabDocs/
Orthomax.html
As for hardware, there are specialized systems that will Externally
display a pseudo-3D of stereo pairs,
but I do not believe they actually physically process stereo pairs
into useable elevation maps for GIS data.
Perhaps I am wrong, but I am sure these systems are crippling in cost
anyways and a fully licensed software solution would be much cheeper.
This is pretty much all the informations I have attained a few months
ago in investigating the same end.
So hopefully there is something in here that will guide you to your
final destination point.
Good luck, and I would really love to hear how things develop at your
end.
cheers!
Mars
On 30-Jun-07, at 7:20 PM, Glynn Clements wrote:
>
> Markus Neteler wrote:
>
>>> I am involved in a project using aerial photographs, earlier we have
>>> been using dias and an 'old fashioned' stereoscope for
>>> interpretations.
>>> Now we are to get the pictures as files. What kind of hard- and
>>> software
>>> do we need to view stereoscopic images on screen? Is this possible
>>> within grass / ossim or otherwhere in the free software family?
>>> Do we
>>> need to get some special hardware, or is is possible to calculate
>>> the
>>> heightinformation from a couple of georeferenced images and then
>>> use eg
>>> nviz? Does anyone here have any experience on that kind of work?
>>
>> As far as I know, there is some stereo support in openGL
>
> I don't think that helps here. AFAICT, Morten is interested in
> converting a stereo pair of 2D images to 3D, while OpenGL's stereo
> support amounts to support for rendering to a stereo pair (you still
> have to render each view separately).
>
> --
> Glynn Clements <glynn at gclements.plus.com>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.osgeo.org/pipermail/grass-user/attachments/20070701/82dc8b0d/attachment.html
More information about the grass-user
mailing list