Ghosting

[Syv Ritch]

Hi,

I have a problem with "ghosting". I use 2011.0.0 : +1.67, 0, -1.67 for
each shot and then each section of the panorama. Hugin uses Enfuse (I
think.) The problem is when there are moving people, trees, grass... I
get a blur.

Is there a way of selecting one of the frame? or saying to Enfuse to
use the frame +0 if it doesn't match with the other 2 of +1.67 and
-1.67?

Thanks

[Carlos Eduardo G. Carvalho (Cartola)]

Hi,

you can use excluding or including masks on hugin. There is a "Mask" tab there. You can also generate the separated images setting one of the "Remapped Images" options at the Stitching tab, like "Exposure corrected, low dinamic range", but the option depends on what you are doing. Then you can use the remapped images on your preferred image editor, like GIMP or Photoshop.

Bye,

Carlos E G Carvalho (Cartola)
http://cartola.org/360

2011/8/14 Syv Ritch <elfr...@gmail.com>

--
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 hugi...@googlegroups.com
To unsubscribe from this group, send email to hugin-ptx+...@googlegroups.com
For more options, visit this group at http://groups.google.com/group/hugin-ptx

[kfj]

On 14 Aug., 06:50, Syv Ritch <elfrog...@gmail.com> wrote:
>
> I have a problem with "ghosting". I use 2011.0.0 : +1.67, 0, -1.67 for
> each shot and then each section of the panorama. Hugin uses Enfuse (I
> think.) The problem is when there are moving people, trees, grass... I
> get a blur.

While ghosting can be controlled with masks when blending several
shots with enblend, as far as I know masking source images for enfuse
does not produce good results. The reasons is that for every single
pixel, exposure blending takes information from all images to combine
them into a result. If parts of one source image are masked out, a
discontinuity appears in the output, since this image's masked pixels
are not available, while their unmasked neighbours are used. This is
particularly visible with hugin's simple on/off masks.

If you want to work around the problem, you'll have to repeat your
brackets and run a deghosting routine on sets of images with the same
exposure - and then exposure-fuse the resulting deghosted images. When
it comes to fluttering leaves and rippples on water, this clearly
won't work. For this situation it sometimes works to 'steepen' the
probability functions enfuse uses. Have a look at the discussion in

http://groups.google.com/group/hugin-ptx/browse_thread/thread/b98d6bf1bb78ee31#

this has effects which approach selecting a specific frame, like when
the fluttering leaves and their background are all within a reasonably
narrow range of values and particularly well-exposed in one of the
images.

> Is there a way of selecting one of the frame? or saying to Enfuse to
> use the frame +0 if it doesn't match with the other 2 of +1.67 and
> -1.67?

So, again, as far as I know there isn't. There may be ways of tricking
hugin/enfuse into doing what you want, though, and I'd be curious if
someone comes up here with a method!

Kay