I was wondering how big the errors introduced by
3D features in the assumed 2D target plane were.
So I made the attached diagram.
It appears "obvious" that, assuming the whole FOV is 'R'
pixels (resolution), that the error is a simple proportion,
since the triangle "projected" by the feature
is similar to the FOV triangle of the camera.
Thus; error (in pixels) = (v/L) * R.
This is independent of FOV, which is interesting.
So, for "perfection" (less than 1 pixel error) with my
8 M Pixel camera, (3264 horizontal pixels), from a shooting distance of 24"
(600mm)
1 = v/600 * 3264
v= 600/3264
= 0.2 mm.
Wowza, I wasn't expecting *that*.
Conversely, if we assume a 3mm wrinkle
in the target;
1 = 3/L * 3264
L= 3 * 3264 = 9792mm
I would need to be 10 metres away to get the error out.
Practical conclusion; I need a means to deal with these
errors in post processing, since it is not reasonable
to eliminate them at shooting time.
BugBear
--
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 on the web visit
https://groups.google.com/d/msgid/hugin-ptx/556C2472.1010600%40papermule.co.uk.
For more options, visit https://groups.google.com/d/optout.
