Area in focus

[mau]

Dear all,

I wonder if it's possible to automatically detect (and maybe mark) the area which is reasonably well in focus of an image, with digikam or maybe with other tools. Any ideas?

Thank you for your inputs.
B.M.
_______________________________________________
Digikam-users mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-users

[tosca]

I think darktable has something like that (I use the program, but not this functionality), but it works with RAW pictures only.

Marie-Noëlle

2015-10-01 10:42 GMT+02:00 B. M. <[hidden email]>:

Dear all,

I wonder if it's possible to automatically detect (and maybe mark) the area which is reasonably well in focus of an image, with digikam or maybe with other tools. Any ideas?

Thank you for your inputs.
B.M.
_______________________________________________
Digikam-users mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-users

--

_______________________________________________
Digikam-users mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-users

[Gilles Caulier-4]

We have 2 entries in bugzilla about this topic to look which area have been focused while shot. The first deal is to read metadata where camera store this information. This way is fast.

https://bugs.kde.org/show_bug.cgi?id=138704

There is another entry which require a different approach : analysis the content of image to see where are the most focused pixels. This way is used in Darktable. This way is slow of course because whole pixels images must be parsed. 

https://bugs.kde.org/show_bug.cgi?id=314022

As we have already an algorithm to detect blur image, the second point will be easy to implement i think. For the first one, Exiv2 must be patched to decode camera marker-notes relevant of focus point. As each camera maker use a different approach, this will be complex. Note : these information are stored in private Exif area and of course are not documented.

Both way can be proposed for students next summer...

Gilles Caulier

2015-10-01 10:52 GMT+02:00 Marie-Noëlle Augendre <[hidden email]>:

I think darktable has something like that (I use the program, but not this functionality), but it works with RAW pictures only.

Marie-Noëlle

2015-10-01 10:42 GMT+02:00 B. M. <[hidden email]>:

Dear all,

I wonder if it's possible to automatically detect (and maybe mark) the area which is reasonably well in focus of an image, with digikam or maybe with other tools. Any ideas?

Thank you for your inputs.
B.M.
_______________________________________________
Digikam-users mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-users

--

   

et aussi sur Facebook :

• Agir pour le futur
• Vallée Borgne Info
• Photographe en Cévennes
  

_______________________________________________
Digikam-users mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-users

_______________________________________________
Digikam-users mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-users

[Anders Lund]

Torsdag den 1. oktober 2015 11:11:09 skrev Gilles Caulier:

> We have 2 entries in bugzilla about this topic to look which area have been
> focused while shot. The first deal is to read metadata where camera store
> this information. This way is fast.
>
> https://bugs.kde.org/show_bug.cgi?id=138704
>
> There is another entry which require a different approach : analysis the
> content of image to see where are the most focused pixels. This way is used
> in Darktable. This way is slow of course because whole pixels images must
> be parsed.

The image data probably contains the focus point(s) active, but in case of
recomposing they may not point to the actual focused area, which favors the
slower analysis. Nice to have, though :)

>
> https://bugs.kde.org/show_bug.cgi?id=314022
>
> As we have already an algorithm to detect blur image, the second point will
> be easy to implement i think. For the first one, Exiv2 must be patched to
> decode camera marker-notes relevant of focus point. As each camera maker
> use a different approach, this will be complex. Note : these information
> are stored in private Exif area and of course are not documented.
>
> Both way can be proposed for students next summer...
>
> Gilles Caulier
>
> 2015-10-01 10:52 GMT+02:00 Marie-Noëlle Augendre <[hidden email]>:
> > I think darktable has something like that (I use the program, but not this
> > functionality), but it works with RAW pictures only.
> >
> > Marie-Noëlle
> >
> > 2015-10-01 10:42 GMT+02:00 B. M. <[hidden email]>:
> >> Dear all,
> >>
> >> I wonder if it's possible to automatically detect (and maybe mark) the
> >> area which is reasonably well in focus of an image, with digikam or maybe
> >> with other tools. Any ideas?
> >>
> >> Thank you for your inputs.
> >> B.M.
> >> _______________________________________________
> >> Digikam-users mailing list
> >> [hidden email]
> >> https://mail.kde.org/mailman/listinfo/digikam-users
> >
> > --
> > [image: Agir pour le futur] <http://agirpourlefutur.com>    [image:
> > Vallée Borgne Info] <http://vallee-borgne.info>
> > <http://marie-noelle-augendre.com>[image: Cévennes Plurielles]
> > <http://cevennes-plurielles.com>
> >
> > et aussi sur Facebook :
> >    - Agir pour le futur
> >    <https://www.facebook.com/AgirPourLeFutur?ref=bookmarks>
> >    - Vallée Borgne Info
> >    <https://www.facebook.com/valleeborgne.info?ref=bookmarks>
> >    - Photographe en Cévennes
> >    <https://www.facebook.com/PhotographeEnCevennes?ref=bookmarks>
> >
> > _______________________________________________
> > Digikam-users mailing list
> > [hidden email]
> > https://mail.kde.org/mailman/listinfo/digikam-users

_______________________________________________
Digikam-users mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-users