On 24 Nov., 19:05, A319 <[email protected]> wrote:
> We took about 150 pictures
> along 10 parallel routes (10 series of pictures) out of the window.

Looking at your images, I notice that they are quite distorted. It
looks as if you had actually taken them through a window, and probably
a not very plane one. With starting material like that, the control
point generators are having a very hard time. If you have to take
aerial photogaphs through a window, make sure you always hold the
camera as close to the pane as possible and set the camera to infinite
distance (pocket cameras often have a little 'mountains' symbol for
that). If there is a pane, though, and you try to photograph
downwards, you hit the pane at a shallow angle, creating bad
distortions. Ideally you'd have the camera pointing straight down
without a pane in between; I realize this may not be feasible, but
with a pane in between your images become very hard to process at all.
What I don't understand about your images is why they are all
different sizes. This confuses the optimizer, since it tries to create
different fields of view etc. for every image. Best to leave them as
you get them out of the camera - if there are any artifacts in them
(plane's wings etc.) it's easy to exclude them using the masking
feature.
If at all possible, all the images should be taken with similar yaw
and pitch, to allow initial optimization of X,Y and Z  inside the
strips with the pitch and yaw set to estimated figures. So if you can,
just mount the camera and leave it like that. If you have a camera
that you can remote-control, you can mount it on the plane pointing
straight down and control it from inside the plane. Try using short
exposure times (if you can't set the exposure time, use a high ISO
setting) to minimize distortions due to vibrations.
Finally, you should take the images in regular time intervals and try
to create sufficient overlap between them, I found your images don't
always overlap sufficiently.

> As you can see, there are serval problems with the highway. What you
> can not see are errors in the forest.

I see the biggest problem in the pictures. I doubt you'll find any
tool to get a decent output from that batch. With a lot af handwork
you'll approach a solution, but it will be far from optimal. I
recommend you do another take.

with regards
Kay

-- 
You received this message because you are subscribed to the Google Groups 
"Hugin and other free panoramic software" group.
A list of frequently asked questions is available at: 
http://wiki.panotools.org/Hugin_FAQ
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at http://groups.google.com/group/hugin-ptx

Reply via email to