Harry -
On Apr 10, 7:30 pm, Harry van der Wolf <[email protected]> wrote:
> --- snip ---
> I also use enfuse quite often for noise reduction, both for normal photo's,
> bracketed sets and darker panorama's.
>
> I can't find any documentation related to that on the web, not even on the
> enblend/enfuse project home page [4], so I assume it is a little known or
> even unknown fact.
The very document you are citing, namely the new
Enfuse documentation features a section on noise
reduction called "6.2 Repetition – Noise Reduction":
http://wiki.panotools.org/Enfuse_reference_manual#Repetition_.E2.80.93_Noise_Reduction
And there is a section on Enfuse's weighting
algorithm, where section "3.1.1 Weighted
Average" in particular mentions the limiting
case of an "arithmetic average".
> --- snip ---
> I assume this enfuse functionality is an undiscovered gem as it has never
> been the purpose of the enfuse tool. It's just a nice side effect, and a
> very effective one.
Prior art.
> I think it is very nice functionality. I know most camera's nowadays have
> noise reduction systems but I find these inferior to enfuse.
It is not quite fair to compare the noise
reduction of a camera that works on just one
image with the multi-image technique you put to
work with Enfuse.
Moreover, the in-camera noise reduction systems
are not necessarily based on averaging or
k-selection (read: median) algorithms.
Dark-frame subtraction is common, though.
> I assume you can even improve the result with the enfuse expert options,
> but this is also an area I'm just beginning to discover.
To my knowledge there is not even an Expert
Option to improve on the averaging.
We have several possibilities to get you closer
to the cigar.
(1) Offer different exposure weight functions.
This one is easy. On the command line, I can
imagine something like
--wExposure=1.0:gaussian
--wExposure=1.0:double-exponential
--wExposure=1.0:distance
where the part after the colon selects the
weighting function over the normalized luminance
interval. Internally, the weighting functions
already are functors...
(2) Offer different weight contribution functions.
Currently, we use "--SoftMask" to let contribute _every_
pixel according to its weight ("democracy" -- well,
kind of), and "--HardMask" where only the pixel
with the largest weight contributes
("monarchy").
As you might know, I plan an extension for the
staging branch of Enblend/Enfuse
https://code.launchpad.net/~cspiel/enblend/staging
that lets the user choose between hard mask and
soft mask with a (quasi-)continuous parameter.
I don't know whether we need (1), but hey, this
is what "staging" is intended for: experiments,
tests, and improvements.
Best,
Chris
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups
"hugin and other free panoramic software" group.
A list of frequently asked questions is available at:
http://wiki.panotools.org/Hugin_FAQ
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at http://groups.google.com/group/hugin-ptx
-~----------~----~----~----~------~----~------~--~---
