BugBear wrote on Thu, 23 Oct 2014 at 13:22:40 +0100
Oh, and I did the optimisation "mosaic style".
I left the base image locked, used the same lens for both, and optimised
the other image
for roll, X Y Z, having set only TWO control points (on the corners of
the upper edge of the lower map, and the lower edge of the upper map).
This (obviously?) gave a 0 error solution.
---------------------
Sorry, I hadn't understood what you were doing, because I gave an
example applicable to a central viewpoint, whereas I now understand you
are trying to superimpose two photos of (I assume) the same original,
which is a map. Therefore the mosaic mode that depends on X, Y and Z
does seem the one to use.
On the scaling point that was the subject of your previous message, I
can only say (1) in my case at least this does not seem to be a question
of the scaling value set in preferences for the Assistant, because I
have never changed it from its default of 70% and yet there was no
scaling in the example I tried, and (2) in one instance I included a
zoomed-in image of a troublesome area and clicking the Calculate optimal
size button gave me a colossal total - as far as I could tell it was
based on the very high resolution of the zoomed-in area, but treated as
applying to the whole mosaic. On the face of it the stitcher calculates
the value that retains the highest resolution present. But perhaps
someone can give a more informed account of how Calculate optimal size
works.
I have superimposed maps, but I think always they were different maps of
the same area that I wanted to compare - a tithe map with a modern
Ordnance Survey map, for instance. I've normally just scaled, rotated
and shifted one relative to the other within the GIMP, though I think I
have used a panorama program to do the calculations for me. At that
stage I used the small-angle approximation that assumes rays from the
different points within the images are effectively parallel. Now I
would certainly try the translation mode.
As to scaling, personally I would not feel that the advantages of
retaining the original version justified the effort in getting the
second image to match it exactly - in none of my map images does the
detail seem that well-defined in the first place. If Calculate optimal
size gave an unhelpful total I would estimate what I think it should be
and insert that in the stitcher tab. You need only insert one value, in
the canvas width or height, and the other values are calculated in
proportion. But I think you have to base your calculation on the
cropped values. Then I would use the two images produced by the
stitcher, which I would know are already aligned.
Roger Broadie
--
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/54495A27.5010000%40ogea.freeserve.co.uk.
For more options, visit https://groups.google.com/d/optout.
