[Space] Photo uploads
cm.hardware.software.elsewhere at gmail.com
Mon Feb 8 15:31:19 PST 2010
On Mon, 2010-02-08 at 14:42 -0800, Blake Barrett wrote:
> Check out panotools: http://panotools.sourceforge.net/
I've stumbled over this last night ... not sure if it takes longer to
understand how to install and interface panotools, or how to come up
with some Mathematica magic.
I tend to be fast inventing wheels and slow using others'.
> On Mon, Feb 8, 2010 at 12:31 PM, Christie Dudley <longobord at gmail.com>
> What would we need to do to get a calibration picture? You
> mentioned something in IRC about it, but I'm not familiar with
> what would be needed in order to correct the fisheye. We can
> probably take care of that on Tuesday when we do our recap.
> "The road of excess leads to the palace of wisdom." -- W.
In any case, a well defined rectangular (or better yet, square) grid on
a flat surface (e.g., a whiteboard), filling the entire field of view,
at a defined distance from the camera lens
(see the pictures at the top of
would be a good start.
Maybe another picture of the grid with a handful of markers that allow
to determine absolute postition could help, too.
>From a theoretical point of view, a random picture with a delta function
autocorrelation under translation, rotation, shearing, etc., would be
very helpful, but this reference picture is hard to construct, and would
need to be available as digital image, too, and would have to be
projected on a flat surface without projection errors - which makes it
difficult practically if the picture is too big to fit on a flat screen.
A regular or irregular, but well defined array of dots might be even
better as reference picture than a grid.
> The outer bounds is only the beginning.
> On Mon, Feb 8, 2010 at 1:27 AM, Christoph Maier
> <cm.hardware.software.elsewhere at gmail.com> wrote:
> On Mon, 2010-02-08 at 01:13 -0800, Ed Gutman wrote:
> > +1 on the congrats and +1 on being a lurker.
> This is, indeed, noisebridge's highest achievement so
> Technical question:
> What is the distortion function of the wide angle lens
> It would be interesting to see how apparent the
> Earth's curvature really
> is at FL601.
> If the distortion function is not available:
> Could someone post a calibration picture of a square
> ruled flat surface?
> lurking from San Diego and looking forward to hearing
> more of this next
> weekend at BIL
More information about the Space