 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Hey, Unka Ken (or anyone else).
I find myself in need of software that can take a stereo pair and generate
a credible heightfield from it. It doesn't have to be perfect; I expect to
have to do some editing. I know such software exists, because it's used to
create DEMs, but I've never seen any available to the public. If anyone
knows of such software, particularly if it's free, please please let me
know. Thanks!
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Ron Parker wrote:
>
> Hey, Unka Ken (or anyone else).
>
> I find myself in need of software that can take a stereo pair and generate
> a credible heightfield from it. It doesn't have to be perfect; I expect to
> have to do some editing. I know such software exists, because it's used to
> create DEMs, but I've never seen any available to the public. If anyone
> knows of such software, particularly if it's free, please please let me
> know. Thanks!
Unka Ken gotta no clue. Neer heara such a thinga that. Whatcha wanna for anyway ?
--
Ken Tyler
mailto://tylereng@pacbell.net
http://home.pacbell.net/tylereng/links.htm
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Tue, 03 Aug 1999 18:50:22 -0700, Ken wrote:
>
>
>Ron Parker wrote:
>>
>> Hey, Unka Ken (or anyone else).
>>
>> I find myself in need of software that can take a stereo pair and generate
>> a credible heightfield from it. It doesn't have to be perfect; I expect to
>> have to do some editing. I know such software exists, because it's used to
>> create DEMs, but I've never seen any available to the public. If anyone
>> knows of such software, particularly if it's free, please please let me
>> know. Thanks!
>
>Unka Ken gotta no clue. Neer heara such a thinga that. Whatcha wanna for anyway ?
You can't see thousands of uses for it just off the top of your head? I
actually found one example of such software, but of course it's an expensive
high-end GIS package. Its official use is to take two or more aerial
photographs and generate a DEM from them, which could then be used as a
heightfield in POV or for more sophisticated uses in watershed analysis or
other high-minded environmental pursuits. My goal is somewhat more
pedestrian: I have a friend who teaches earth science to seventh-graders.
One of their favorite projects is the one where they build clay mountains,
then submerge them to various depths and draw the contour lines. I was
looking for free or inexpensive software that could be used with a digital
camera to automate the process, so they can see how it's done in the digital
age. I figured that if such a thing existed, one or more of you 3d guys
would know where to find it. Unfortunately, I'm beginning to believe that
there is no such thing after all.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Ron Parker wrote:
>
> On Tue, 03 Aug 1999 18:50:22 -0700, Ken wrote:
> >
> One of their favorite projects is the one where they build clay mountains,
> then submerge them to various depths and draw the contour lines.
Sounds like fun...
I was
> looking for free or inexpensive software that could be used with a digital
> camera to automate the process, so they can see how it's done in the digital
> age.
And derive them of the fun?
I figured that if such a thing existed, one or more of you 3d guys
> would know where to find it. Unfortunately, I'm beginning to believe that
> there is no such thing after all.
I know someone who will probably know such a package but it's likely to be same
as the one you found, or similar like in expensive.
I thought it shouldn't be hard to make such a program if you know exactly what
to use it for (straightforward). Then I thought, the problem is probably not the
program (that's just calculating a parallax or what-do-you-call-it), but the
photography. You have to know the exact location of the camera at two points.
Also, you'll probably need quite a high resolution to get a result with a
reasonable amount of details. If the photo-part is not a problem I can't imagine
that the programming-part would be (not for you, would it?).
Regards,
Remco
http://www.xs4all.nl/~remcodek/
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Wed, 04 Aug 1999 16:32:00 +0200, Remco de Korte wrote:
>I was
>> looking for free or inexpensive software that could be used with a digital
>> camera to automate the process, so they can see how it's done in the digital
>> age.
>
>And derive them of the fun?
Not at all. They'd still do it the old way and have the opportunity to
get messy, they'd also get to see another way of doing the same thing,
perhaps with some cool POV visualization and some more advanced watershed
management stuff (flow lines, etc.) thrown in courtesy of GRASS.
>I thought it shouldn't be hard to make such a program if you know exactly what
>to use it for (straightforward). Then I thought, the problem is probably not the
>program (that's just calculating a parallax or what-do-you-call-it), but the
>photography. You have to know the exact location of the camera at two points.
>Also, you'll probably need quite a high resolution to get a result with a
>reasonable amount of details. If the photo-part is not a problem I can't imagine
>that the programming-part would be (not for you, would it?).
High resolution is likely to be necessary, yes. I'm not sure about knowing
the exact location of the camera. Separation distance, distance from the
terrain, and some info about lens geometry might be necessary, true, unless
you were willing to cheat and feed it elevations for the summit and the
lowlands and let it scale its results to fit. I think even USGS cheats by
putting reflective markers at known locations in the area to be photographed,
so that rectification can take place separate from stereocorrelation.
Unfortunately, I did some experiments and it's much more difficult than it
looks to recover the parallax information. And of course I wouldn't want
such a project to take time away from valuable pursuits like the superpatch. :)
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Eh..er..um... there is something which is somewhat like this...
http://www.photomodeler.com
Ron Parker <par### [at] fwi com> wrote in message
news:37a83bbf@news.povray.org...
> On Tue, 03 Aug 1999 18:50:22 -0700, Ken wrote:
> >
> >
> >Ron Parker wrote:
> >>
> >> Hey, Unka Ken (or anyone else).
> >>
> >> I find myself in need of software that can take a stereo pair and
generate
> >> a credible heightfield from it. It doesn't have to be perfect; I
expect to
> >> have to do some editing. I know such software exists, because it's
used to
> >> create DEMs, but I've never seen any available to the public. If
anyone
> >> knows of such software, particularly if it's free, please please let me
> >> know. Thanks!
> >
> >Unka Ken gotta no clue. Neer heara such a thinga that. Whatcha wanna for
anyway ?
>
> You can't see thousands of uses for it just off the top of your head? I
> actually found one example of such software, but of course it's an
expensive
> high-end GIS package. Its official use is to take two or more aerial
> photographs and generate a DEM from them, which could then be used as a
> heightfield in POV or for more sophisticated uses in watershed analysis or
> other high-minded environmental pursuits. My goal is somewhat more
> pedestrian: I have a friend who teaches earth science to seventh-graders.
> One of their favorite projects is the one where they build clay mountains,
> then submerge them to various depths and draw the contour lines. I was
> looking for free or inexpensive software that could be used with a digital
> camera to automate the process, so they can see how it's done in the
digital
> age. I figured that if such a thing existed, one or more of you 3d guys
> would know where to find it. Unfortunately, I'm beginning to believe
that
> there is no such thing after all.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Wed, 4 Aug 1999 11:01:02 -0400, Bill DeWitt wrote:
>Eh..er..um... there is something which is somewhat like this...
>
>http://www.photomodeler.com
I looked at photomodeler yesterday, actually. It looks like a neat
program, but it does two things that don't work for me: it only seems
to work with polygonal data, and it requires me to identify and
correlate the salient features. Stereocorrelation software doesn't
have those limitations.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Well, I am sure you are right in the large scale, and not that I really
understand what you are talking about, but I wonder if for the expedient of
making some models for school children, if it might be made to suffice
somehow.
Barring you actually finding what you really want, of course...
Ron Parker <par### [at] fwicom> wrote in message
news:37a85bdd@news.povray.org...
> On Wed, 4 Aug 1999 11:01:02 -0400, Bill DeWitt wrote:
> >Eh..er..um... there is something which is somewhat like this...
> >
> >http://www.photomodeler.com
>
> I looked at photomodeler yesterday, actually. It looks like a neat
> program, but it does two things that don't work for me: it only seems
> to work with polygonal data, and it requires me to identify and
> correlate the salient features. Stereocorrelation software doesn't
> have those limitations.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Ron Parker heeft geschreven in bericht <37a85246@news.povray.org>...
.....
>High resolution is likely to be necessary, yes. I'm not sure about knowing
>the exact location of the camera. Separation distance, distance from the
>terrain, and some info about lens geometry might be necessary, true, unless
>you were willing to cheat and feed it elevations for the summit and the
>lowlands and let it scale its results to fit. I think even USGS cheats by
>putting reflective markers at known locations in the area to be photographed,
>so that rectification can take place separate from stereocorrelation.
Imagine a glass plate, on both sides is a painted dot.
Set up the cameras and set focus between both dots, this is the average object
distance (Od). Measure the distance between the two cameras, centere of lens to
centre of lens. This is the StereoBase (Sb). You also need the distance from the
lens to the film plane (Id). Next, measure the parallax between the points in
your pix. Then multiply the parallax by Od^2/Sb*ID. This gives the depth between
the two dots on the glass (thickness of the glass)
Biggest problem for the software is to find the corresponding pixels in the two
pictures. I've done this lots of times by hand. I use a kite plus sterocamera to
do some aerial photography.
http://rst.gsfc.nasa.gov/Start.html
This is a site on remote sensing, I have not explored it yet but chapter 11 is
about stereo stuff.
ingo
--
Met dank aan de muze met het glazen oog.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On Wed, 4 Aug 1999 19:28:33 +0200, ingo wrote:
>Ron Parker heeft geschreven in bericht <37a85246@news.povray.org>...
>Biggest problem for the software is to find the corresponding pixels in the two
>pictures. I've done this lots of times by hand. I use a kite plus sterocamera to
>do some aerial photography.
That's the part I'm concerned about, actually. I can generally figure out the
rest from scratch when I need it (That's not to say that I won't just use what
you've derived instead... I will.) I assume that real-world software does this
by sliding a window across the right image and doing some sort of statistical
correlation with a fixed window on the left image, but I haven't been able to
create anything here that gives me satisfactory results. That may be due to
a lack of good data, however. I will keep trying, of course. I won't actually
need any of this stuff for at least six months, and if I don't have it the
world won't end.
>http://rst.gsfc.nasa.gov/Start.html
>This is a site on remote sensing, I have not explored it yet but chapter 11 is
>about stereo stuff.
Thanks, That's a very nice site in and of itself, though I didn't find anything
I could use at the moment. Nevertheless, I will be passing it along to my
teacher friend to add to the website she's making for her students as a
starting point for independent study.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
|
|
