Hello Derek,
I am attempting to stitch a large and fairly messy set of aerial photos,
and not having much luck. That's not a huge surprise because the photos
were hand-shot out of a moving aircraft, so they were certainly not taken
with this application in mind. However, because the area being photographed
is coastal barrier islands, which are fairly flat, I was hoping I might
nonetheless be able to get something that appears reasonable (even if it is
not the same as a georeferenced map image).
I agree that this should be possible. Manually removing control points
from moving waves and anything higher than sea level may help (but I
didn't find it necessary).
However, whenever I get beyond 4-6 images, the optimization generally seems
to break down and I get ridiculous results or the panorama fails to stitch
at all. I am scripting this due to the large number of images, and my flow
is as follows:
1. pto_gen -o {pto_file} {images}
2. cpfind --celeste --multirow --cache --fullscale -o {pto_file} {pto_file}
3. pto_var --opt y,p,r,TrX,TrY,TrZ --anchor={middle_image} -o {pto_file}
{pto_file}
4. autooptimiser -n -o {pto_file} {pto_file}
5. pano_modify --projection=0 --fov=AUTO --canvas=AUTO --crop=AUTO
--ldr-file=JPG --ldr-compression=99 -o {pto_file} {pto_file}
6. nona -m TIFF_m -o {prefix} {pto_file}
7. enblend --compression=99 -o {output_name} {prefix}*.tif
The images are public, so in case I can convince anyone more knowledgeable
than me to give this a try, I'm using the Raw images from
here:
https://experience.arcgis.com/experience/010503b4c64b4ff6a7f3570220a53647/page/Data-Explorer-
Specifically, I've been experimenting with "21May2018
Camera2-[6290-6325].jpg"
For easier download:
https://twi-aviandata.s3.amazonaws.com/HighResolutionImages/2018/21May2018%20Camera1-${i}.jpg
I agree that it is not a simple example. I'm getting this:
http://78.46.190.157:8080/birds.jpg, ie, good alignment but the mapping
is a little wonky. While I don't think my result is great I have some
thoughts about that:
-- the images overlap far too much. I have removed 2 out of every 3.
That could be done systematically, if they always overlap that much.
One could also truncate each image.
-- I assume the images have been shot out the side of the plane, ie, not
vertically down. Then it is not correct to set yaw and pitch of the
anchor to zero. (I'm getting a yaw of ~20degrees).
-- freely optimising y/p/r/Tx/Ty/Tz of many images hardly ever converges
into the intended minimum. The obvious strategies are to optimise a few
images, add a few more, optimise them together, etc. Here, I have
optimised first Tx/Ty for all, then y/p/r for all except the anchor,
then y/p for the anchor. Tz has only been included for some images --
optimising it for all yields better agreement but a worse mapping (and I
assume the plane has been at a consistent height). I don't have a good
idea how to deal with that -- the problem is obviously that the degrees
of freedom are not orthogonal.
-- having any right angles on the ground and using them for line control
points would help very much.
best regards, lukas wirz
--
A list of frequently asked questions is available at:
http://wiki.panotools.org/Hugin_FAQ
---
You received this message because you are subscribed to the Google Groups "hugin and other free panoramic software" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to [email protected].
To view this discussion visit
https://groups.google.com/d/msgid/hugin-ptx/e9382a26-412e-4853-85eb-6f4e5dd51e3d%40posteo.net.
