Photos {Was: thunder & hailstones}

On Wed, 28 Sep 2005 21:53:21 +0100, a l l y wrote:

> I think the problem is the same one map-makers have. They are
> looking down on to a sphere (the earth) and trying to map this
> convex shape on a flat surface. I am *inside* a sphere (the earth's
> atmosphere) and trying to make flat images of a concave surface (the
> sky).

Similar but the misalgnment in your photo is due to the camera moving
and not observing from *exactly* the same point. Agreed you can't do
it with out some distortion of the image but you can make the
distortion gradual rather the step. B-)

> You can get away with a simple panorama where the camera stays at
> the same height (I've done this several times fairly successfully)
> but the minute you start panning upwards as well as on the
> horizontal plane things get really complicated. I'm sure there
> must be software that'll fix this, but I don't try it often enough
> to make it worthwhile.

Getting the images to align correctly without fudging helps a great
deal. I've done panoramas as well but the joins are always a bit of a
fudge but that is without a proper mount. It should be possible to
have invisible joins without fudging. If it works in one plane it
should work in the other, probably not both at the same time though. I
think it would work fairly well for the area of even a large rainbow.

--
Cheers new5pam@xxxxxxxxxxx
Dave. pam is missing e-mail



.

Relevant Pages

  • Re: registration of spherical images
    ... of the original images would actually correspond to simple shifts on ... the sphere, but what about roll? ... and then map back onto your sphere. ... tilt but don't know about roll. ...
    (sci.image.processing)
  • Re: registration of spherical images
    ... the sphere, but what about roll? ... Then, once your spherical images have been mapped to the plane, maybe ... and then map back onto your sphere. ...
    (sci.image.processing)
  • Re: registration of spherical images
    ... the sphere, but what about roll? ... Then, once your spherical images have been mapped to the plane, maybe ... spherical projection and then mosaic them. ... tilt but don't know about roll. ...
    (sci.image.processing)
  • 3D Lunar transformation *on Titan*?
    ... >image of the moon is plonked onto a sphere which can be rotated to unfurl ... Have you tried doing this with the raw images just coming in from ... Cassini at Titan? ... if you remap those onto a sphere and rotate, ...
    (sci.astro.amateur)