Improving photos

Hello,

Some other general digital photography topic.

Not all photos (I'd say minority) is coming from camera ideal.
Sometimes you need to remove noise, highlight dark areas, improve
levels.

How often do you use this techniques?
What are you doing with original photos?
What techniques do you use?

m.

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

[Gilles Caulier-4]

2007/4/17, Mikolaj Machowski <[hidden email]>:

Hello,

Some other general digital photography topic.

Not all photos (I'd say minority) is coming from camera ideal.
Sometimes you need to remove noise,

Editor can be used for that with Noise Reduction plugin.
 

highlight dark areas,

Speed way is to  use curves adjust tool and reduce highlight level.

Using RAW file will be better to fix the image duing pixels/bits resolution.

But the highlight is a hard problem with digital camera, because CCD works weird in highlight.

If you have a FUji camera, the supper CCD is a first approach to fight hightlight.

With CCD camera, the High Dynamic Range pictures is the best solution, but only for static subjects.

Of course, a program like Krita witch can work on complex selected area will be better, be require long correction...

improve
levels.

Levels Adjust tool and auto-levels can give good work...
 

How often do you use this techniques?
What are you doing with original photos?
What techniques do you use?

There is a lots of articles on the web based generaly on Photoshop. Google is your friends.

Gilles

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

Mikolaj Machowski wrote:
> Hello,
>
> Some other general digital photography topic.
>
> Not all photos (I'd say minority) is coming from camera ideal.
> Sometimes you need to remove noise, highlight dark areas, improve
> levels.

most images

>
> How often do you use this techniques?
> What are you doing with original photos?
> What techniques do you use?

digikal/showfoto is very handy to this.

usually I try automatic level (color), it's good 50% of the time, if
not I use the gamma slider of the color fix

and of course the "recadrage" (sorry, don't know the english translation)

jdd


--
http://www.dodin.net
Lucien Dodin, inventeur
http://lucien.dodin.net/index.shtml
_______________________________________________
Digikam-users mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-users

[Gilles Caulier-4]

2007/4/17, jdd <[hidden email]>:

Mikolaj Machowski wrote:
> Hello,
>
> Some other general digital photography topic.
>
> Not all photos (I'd say minority) is coming from camera ideal.
> Sometimes you need to remove noise, highlight dark areas, improve
> levels.

most images

>
> How often do you use this techniques?
> What are you doing with original photos?
> What techniques do you use?

digikal/showfoto is very handy to this.

usually I try automatic level (color), it's good 50% of the time, if
not I use the gamma slider of the color fix

and of course the "recadrage" (sorry, don't know the english translation)

Cropping...

Gilles

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

[Gerhard Kulzer]

Am Tuesday 17 April 2007 schrieb Mikolaj Machowski:
> Hello,
>
> Some other general digital photography topic.
>
> Not all photos (I'd say minority) is coming from camera ideal.
> Sometimes you need to remove noise, highlight dark areas, improve
> levels.
>
> How often do you use this techniques?
Level, color and cropping I use on 95% of my images.
Everything that get's published I have to use Gimp or Krita on top of digiKam.

> What are you doing with original photos?
I keep them. Usually I shoot in double mode creating RAW and JPEG in the
camera. Then the RAw is always the original.

> What techniques do you use?
>
> m.
>
Gerhard

--
Hakuna matata
http://www.gerhard.fr

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

[Jakob "Østergaard"]

Mikolaj Machowski wrote:
> Hello,
>
> Some other general digital photography topic.

Just throwing in my 0.02 Euro as well :)

>
> Not all photos (I'd say minority) is coming from camera ideal.

I use the batch RAW->jpeg converter and find that many of my pictures
just look wrong (much too bright or much too dark) after conversion.
Those that come out wrong, I convert manually.

I also manually convert those that need cropping.

> Sometimes you need to remove noise, highlight dark areas, improve
> levels.

I usually avoid having to remove noise, because the results simply
aren't that good - you can remove some noise, but if the image is really
noisy, I prefer shooting a better image to repairing a noisy one. Of
course you can't do that always :)

> How often do you use this techniques?

I manually convert/adjust/crop most of the images I end up keeping.
Maybe 2 out of 3 or so.

Even if the batch converter does a good job, it's common (for me) to
want a little cropping, or a little tweaking of the colours (usually
just white balance).

> What are you doing with original photos?

1) Import RAW (16-bit)
2) Adjust white balance if need be (Ctrl-W)
3) Adjust levels if need be (Ctrl-L)
4) Crop if need be
5) Save as JPEG (90%)

If the image is very underexposed, I find it useful to adjust the
exposure/black-point in the white-balance tool, to get a better exposure
on the image while adjusting the white balance. An under-exposed image
looks (at least to my eyes) colder than it is, so I tend to over-do the
white balance adjustment on dark images if I don't correct the exposure too.

The above covers 95% of what I do with my photos now. I sometimes
convert to B&W, sometimes rotate a few degrees to correct a camera tilt,
  sometimes use noise reduction or image restoration, but these cases
are rare.

Most of my images are birds in flight (taken with a 300mm prime) or
landscape/nature (taken with a poor 18-55mm zoom which I hope to replace
soon). People who take other kinds of images may well use very different
parts of digikam, I don't know  :)

--
Best regards,
    Jakob Oestergaard
    [The SysOrb Team]

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

[Jürgen Flosbach dk]

Mikolaj Machowski wrote:

> Hello,
>
> Some other general digital photography topic.
>
> Not all photos (I'd say minority) is coming from camera ideal.
> Sometimes you need to remove noise, highlight dark areas, improve
> levels.
>
> How often do you use this techniques?
> What are you doing with original photos?
> What techniques do you use?

Just my way.
------------

My objects are :
Landscape, flowers, macros, family, traveling

After shooting the images

create appropriate directory
create directory underneath called "original"
copy all new images into the "original" directory

delete bad images
create descriptions and tags for the good ones
rename, with the batch tool, all remaining images
    according to my naming scheme.
adjust most of my images and save the result in the
    directory above the "original"

adjusting tasks are :
rotate
crop
level
refocus

For more critical tasks I edit the pictures with Gimp ( unfortunately
Gimp doesn't ) keep the exif information.

Currently I shoot my images as JPEGs in highest resolution but maybe I
go back to shoot in RAW again.

I'm currently reading information about HDI and DRI.
The following link might be of interest :
http://turtle.as.arizona.edu/jdsmith/exposure_blend.php

--
With regards

Juergen Flosbach

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

[Daniel Bauer-2]

On Dienstag, 17. April 2007, Mikolaj Machowski wrote:
> Hello,
>
> Some other general digital photography topic.
>
> Not all photos (I'd say minority) is coming from camera ideal.
> Sometimes you need to remove noise, highlight dark areas, improve
> levels.

I use(d?) UFraw to convert the CR2's. I like the control possibilities UFraw
offers. I must confess that I haven't tried the latest digikam in this
regard, so maybe this point could change in future...

In 99,99999% of all cases I use 8bit (publishers anyway scale down 16bit
images to 8 before printing, so what should I grapple with much larger
files :-)  I use 16bit only, if there are difficult parts in the shadows or
lights, so I can get out more of it.

Although digikam offers possibilities to scale images I always use imageMagick
from the command line. I have a text file on my desktop with the commands I
often use, so I don't have to remember them. I also use imagemagick to strip
metadata from the images if not needed.

All the rest I do with digikam. It is very rare that I have to treat only
parts of a picture (then I have to use gimp - a program that can
do "everything" but is [sorry!] just horrible in the handling). I *love* the
cropping and the text tools of digikam, they are way more comfortable and
faster than what I was used from photoshop before (I didn't have the latest
versions though...). I also like the black-and-white filters of digikam and
use them quite often. For montages, of course, I need to use the Gimp.

> How often do you use this techniques?

always when working digitally.

> What are you doing with original photos?

- CR2's remain untouched in a folder.
- same for converted raw files.
- same for finished photos (in original size)

these folders are copied on another harddisk in same computer plus on other
hard disk in another computer plus on CD's that are stored in a fireproof
save (yes, I am kind of paranoid in this regard...). I never delete any
originals because I know that years later my opinion about what's good or bad
can be the contrary of what I thought before.

Then, starting from "finished photos", but always working with copies, there
are many folders with selections per purpose of use, model, customer, size
etc.

Although all folders are in the digikam album path, I use a SQL database
with "home made" php scripts to organize the archive, as I am not happy with
the current implementation of "digikam-tags=IPTC metadata"....

regards

Daniel
--
Daniel Bauer photographer Basel Switzerland
professional photography: http://www.daniel-bauer.com
Madagascar special: http://www.sanic.ch
_______________________________________________
Digikam-users mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-users

[Gilles Caulier-4]

In 99,99999% of all cases I use 8bit (publishers anyway scale down 16bit
images to 8 before printing, so what should I grapple with much larger
files :-)  I use 16bit only, if there are difficult parts in the shadows or
lights, so I can get out more of it.

I would just to give my viewpoint about shadows/highlight with 16 bits color depth.

16 bits color depth want mean than each pixels have a bigger color resolution than 8 bits. But this is not want mean than you will enlarge the CCD dynamic range ! It still the same especially with highlight.

Of course the 16 bits color depth is better to process color adjustments for the light properlly exposed.

The CCD are different than Human eyes and do not restitute the highlight properlly. This is why if you want to render an image in the same way than human eyes, you need to do High Range picture, using bracketing with camera. Typically like this :

- One picture under exposed to get HighLight
- One picture properlly exposed to get MidLight
- One picture over exposed to get ShadowLight

... and afer, you merge all in the same picture, using a specific format like OpenEXR  for ex.

This is explained certainly better than mine in wikipedia :

http://en.wikipedia.org/wiki/High_dynamic_range_imaging

And the imagering technic is called Tone Mapping. look here :

http://en.wikipedia.org/wiki/Tone_mapping

Like you can see, the HDR pictures are really  really nice, but of course this technic can be only used with static subjects. Generally, we using a tripod during camera shots.

In the future, i would to have a plugin to process HDR image in digiKam using a serie of bracketed shots. It's not very complicated to do.

Note than SuperCCD from Fuji is the only solution to improve the dynamic range of digital camera :

http://en.wikipedia.org/wiki/Super_CCD

... and you can find some explainations of Dynamic Range here :

http://www.dpreview.com/learn/?/Glossary/Digital_Imaging/dynamic_range_01.htm

.. and the relation with famous histogram...


Gilles
 

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

Gilles Caulier wrote:

> The CCD are different than Human eyes and do not restitute the highlight
> properlly. This is why if you want to render an image in the same way
> than human eyes, you need to do High Range picture, using bracketing
> with camera. Typically like this :

for french readers, there is an excellent article of "Chasseur
d'images" this month on the very subject

jdd

--
http://www.dodin.net
Lucien Dodin, inventeur
http://lucien.dodin.net/index.shtml
_______________________________________________
Digikam-users mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-users

Hi,
>
> In the future, i would to have a plugin to process HDR image in
> digiKam using a serie of bracketed shots. It's not very complicated to
> do.
>

This make me think of an idea I had. I don't know if it is something
that has already been done and if it is possible at all.

Usually for a portrait, it is nice to have a blurred background.  This
can be achieved by increasing the aperture using expensive lenses.
A cheaper way to obtain the blurred background is to apply a gaussian
blur on the selection of the background of the picture using  gimp or
krita. But the selection of the background is a difficult and tedious task.
Here is an example :
http://www.outbackphoto.com/workflow/wf_51/essay.html

I think it would be nice to have a way to create automatically the
selection (the depth map).

Maybe one could take several pictures of the scene with the smallest
possible depth of field using different focus distances, then the
pictures are compared to detect the part which are blurred or not. From
this information one could  infer the distance.
At the end this could lead to a tool to select the depth of field
afterwards.

Does anyone has ever seen a software or plugin to do such a thing ?

my 2 cents,

Julien

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

[Gerhard Kulzer]

Am Wednesday 18 April 2007 schrieb Jürgen Flosbach dk:

> Mikolaj Machowski wrote:
> > Hello,
> >
> > Some other general digital photography topic.
> >
> > Not all photos (I'd say minority) is coming from camera ideal.
> > Sometimes you need to remove noise, highlight dark areas, improve
> > levels.
> >
> > How often do you use this techniques?
> > What are you doing with original photos?
> > What techniques do you use?
>
> Just my way.
> ------------
>
> My objects are :
> Landscape, flowers, macros, family, traveling
>
> After shooting the images
>
> create appropriate directory
> create directory underneath called "original"
> copy all new images into the "original" directory
>
> delete bad images
> create descriptions and tags for the good ones
> rename, with the batch tool, all remaining images
>     according to my naming scheme.
> adjust most of my images and save the result in the
>     directory above the "original"
>
> adjusting tasks are :
> rotate
> crop
> level
> refocus
>
> For more critical tasks I edit the pictures with Gimp ( unfortunately
> Gimp doesn't ) keep the exif information.

can be restored like this:
exiftool -tagsfromfile orig.jpg gimp.jpg
>
> Currently I shoot my images as JPEGs in highest resolution but maybe I
> go back to shoot in RAW again.
>
> I'm currently reading information about HDI and DRI.
> The following link might be of interest :
> http://turtle.as.arizona.edu/jdsmith/exposure_blend.php

Gerhard
--
Hakuna matata
http://www.gerhard.fr
_______________________________________________
Digikam-users mailing list
[hidden email]
https://mail.kde.org/mailman/listinfo/digikam-users

[Marcel Wiesweg]

> The CCD are different than Human eyes and do not restitute the highlight
> properlly. This is why if you want to render an image in the same way than
> human eyes, you need to do High Range picture, using bracketing with
> camera. Typically like this :
>
> - One picture under exposed to get HighLight
> - One picture properlly exposed to get MidLight
> - One picture over exposed to get ShadowLight
>
> ... and afer, you merge all in the same picture, using a specific format
> like OpenEXR  for ex.
>

I recently found this one, but I have not yet tried it:
http://qtpfsgui.sourceforge.net/
It's Qt4, and seems to be actively maintained. I think they have incorporated
the source of the various command line tools (pfstools etc.)

If you dont have a tripod but shoot at free hands, things may be slightly
displaced. A few days ago I read about this, also pretty interesting:
http://cyrilleberger.blogspot.com/2007/04/aligning-images.html

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

[Milan Zamazal]

>>>>> "JN" == Julien Narboux <[hidden email]> writes:

    JN> Usually for a portrait, it is nice to have a blurred background.
    JN> This can be achieved by increasing the aperture using expensive
    JN> lenses.

It can be achieved by increasing the aperture using cheap old manual
lenses too.  (But other reasons why software blurring may be useful
still apply.)

Regards,

Milan Zamazal

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