SD9 color adjustment techniques

[Ladehaug]

Hello everyone,

I've spent considerable time researching the color characteristics of the SD9. As a result of this I've made several hue adjustment profiles that can be used in Photoshop to tweak the colors to a more accurate value. Not all pictures need color correction though, it appears to vary with the colors involved, light conditions and the general exposure level.

I'm planning to construct a web page, demonstrating some effective Photoshop techniques/workflows of adjusting the colors. This will include access to my recent hue correction profiles for use with the SD9, and how to use them in the most effective manner.

How many of you would be interested in such a web page?

I also want to create a small gallery of example images that includes one or more versions of the images with adjusted colors. Side by side comparisons will demonstrate the effect of adjusting the colors. In connection with this I'll need permission from you SD9 owners out there to use one or more of your sample images as example images on my web page. I don't own a SD9 myself yet! Those of you who want to give me a permission to use their images, please notify me, either by email or by responding to this post. I won't be using images from users that have not given me a permission. Of course, the images will only be used on my web page and will not be used for anything else than pure examples on this specific page.

I have one last request for those SD9 owners who want to help me a bit further in the study of the SD9 color characteristics. Below is an test image (1024x768) that I want you to make a monitor screenshot of. Make sure the room is dark and use average metering. Set white balance to auto (One can always change this in the RAW file later, but if auto was used, then at least the auto reading will be present in the RAW file). Make 4 additional versions with -2ev, -1ev, +1ev and +2ev. Shoot at smallest resolution (or medium). I don't need full sized images, it's more important that I can have the X3F files. The files can be mailed to me. The test pictures won't be published, they will only be used by me to study the hue color balance.

I would appreciate any comments and/or suggestions!

Regards,
Geir

Test image:

 

[Laurence Matson]

Gosh, this is fun. More homework.

I am interested and would be willing to pay a bit for the tools. Or I'll see what I can do to help you get an SD9 reasonably priced.

And I appreciate this. Constructive efforts move things forward.

Laurence

Hello everyone,

I've spent considerable time researching the color characteristics
of the SD9. As a result of this I've made several hue adjustment
profiles that can be used in Photoshop to tweak the colors to a
more accurate value. Not all pictures need color correction though,
it appears to vary with the colors involved, light conditions and
the general exposure level.

I'm planning to construct a web page, demonstrating some effective
Photoshop techniques/workflows of adjusting the colors. This will
include access to my recent hue correction profiles for use with
the SD9, and how to use them in the most effective manner.

How many of you would be interested in such a web page?

I also want to create a small gallery of example images that
includes one or more versions of the images with adjusted colors.
Side by side comparisons will demonstrate the effect of adjusting
the colors. In connection with this I'll need permission from you
SD9 owners out there to use one or more of your sample images as
example images on my web page. I don't own a SD9 myself yet! Those
of you who want to give me a permission to use their images, please
notify me, either by email or by responding to this post. I won't
be using images from users that have not given me a permission. Of
course, the images will only be used on my web page and will not be
used for anything else than pure examples on this specific page.

I have one last request for those SD9 owners who want to help me a
bit further in the study of the SD9 color characteristics. Below is
an test image (1024x768) that I want you to make a monitor
screenshot of. Make sure the room is dark and use average metering.
Set white balance to auto (One can always change this in the RAW
file later, but if auto was used, then at least the auto reading
will be present in the RAW file). Make 4 additional versions with
-2ev, -1ev, +1ev and +2ev. Shoot at smallest resolution (or
medium). I don't need full sized images, it's more important that I
can have the X3F files. The files can be mailed to me. The test
pictures won't be published, they will only be used by me to study
the hue color balance.

I would appreciate any comments and/or suggestions!

Regards,
Geir

Test image:
http://home.online.no/~ladehaug/dpreview/huetest.png

--
Go Alinghi

--------------------
Current Standings
Alinghi 3
Wind 4
Team New Zealand 0
Kiwi Fans 30

http://www.pbase.com/lmatson/sd9_images

 

[KIM MIN-SU]

More condition need such as

(1) Color temperature of the monitor
(2) Bright/Contrast Level
(3) Graphic Card if required (Color temperature is more flexible)

And I'll send the sample images to you, Can you send me your
e-mail address?

My address : [email protected]

Thanks

Hello everyone,

I've spent considerable time researching the color characteristics
of the SD9. As a result of this I've made several hue adjustment
profiles that can be used in Photoshop to tweak the colors to a
more accurate value. Not all pictures need color correction though,
it appears to vary with the colors involved, light conditions and
the general exposure level.

I'm planning to construct a web page, demonstrating some effective
Photoshop techniques/workflows of adjusting the colors. This will
include access to my recent hue correction profiles for use with
the SD9, and how to use them in the most effective manner.

How many of you would be interested in such a web page?

I also want to create a small gallery of example images that
includes one or more versions of the images with adjusted colors.
Side by side comparisons will demonstrate the effect of adjusting
the colors. In connection with this I'll need permission from you
SD9 owners out there to use one or more of your sample images as
example images on my web page. I don't own a SD9 myself yet! Those
of you who want to give me a permission to use their images, please
notify me, either by email or by responding to this post. I won't
be using images from users that have not given me a permission. Of
course, the images will only be used on my web page and will not be
used for anything else than pure examples on this specific page.

I have one last request for those SD9 owners who want to help me a
bit further in the study of the SD9 color characteristics. Below is
an test image (1024x768) that I want you to make a monitor
screenshot of. Make sure the room is dark and use average metering.
Set white balance to auto (One can always change this in the RAW
file later, but if auto was used, then at least the auto reading
will be present in the RAW file). Make 4 additional versions with
-2ev, -1ev, +1ev and +2ev. Shoot at smallest resolution (or
medium). I don't need full sized images, it's more important that I
can have the X3F files. The files can be mailed to me. The test
pictures won't be published, they will only be used by me to study
the hue color balance.

I would appreciate any comments and/or suggestions!

Regards,
Geir

Test image:

--
SD9 with 24-70DG2.8

 

[PapaRazi]

You've got mail.

We all need a State of Grace

 

[Kendall Helmstetter Gelner]

Hello everyone,

I'm planning to construct a web page, demonstrating some effective
Photoshop techniques/workflows of adjusting the colors. This will
include access to my recent hue correction profiles for use with
the SD9, and how to use them in the most effective manner.

How many of you would be interested in such a web page?

I'd be interested - I don't have an SD9 myself yet either, but will probably be buying on in about a month or so...

--
---> Kendall

 

[Rytterfalk]

Alright!! Now we talkin'n!

And yes, you can use whatever you want. And I'll try to do this screenshot test, but I'm not so sure about my screens calibration,

Hello everyone,

I've spent considerable time researching the color characteristics
of the SD9. As a result of this I've made several hue adjustment
profiles that can be used in Photoshop to tweak the colors to a
more accurate value. Not all pictures need color correction though,
it appears to vary with the colors involved, light conditions and
the general exposure level.

I'm planning to construct a web page, demonstrating some effective
Photoshop techniques/workflows of adjusting the colors. This will
include access to my recent hue correction profiles for use with
the SD9, and how to use them in the most effective manner.

How many of you would be interested in such a web page?

I also want to create a small gallery of example images that
includes one or more versions of the images with adjusted colors.
Side by side comparisons will demonstrate the effect of adjusting
the colors. In connection with this I'll need permission from you
SD9 owners out there to use one or more of your sample images as
example images on my web page. I don't own a SD9 myself yet! Those
of you who want to give me a permission to use their images, please
notify me, either by email or by responding to this post. I won't
be using images from users that have not given me a permission. Of
course, the images will only be used on my web page and will not be
used for anything else than pure examples on this specific page.

I have one last request for those SD9 owners who want to help me a
bit further in the study of the SD9 color characteristics. Below is
an test image (1024x768) that I want you to make a monitor
screenshot of. Make sure the room is dark and use average metering.
Set white balance to auto (One can always change this in the RAW
file later, but if auto was used, then at least the auto reading
will be present in the RAW file). Make 4 additional versions with
-2ev, -1ev, +1ev and +2ev. Shoot at smallest resolution (or
medium). I don't need full sized images, it's more important that I
can have the X3F files. The files can be mailed to me. The test
pictures won't be published, they will only be used by me to study
the hue color balance.

I would appreciate any comments and/or suggestions!

Regards,
Geir

Test image:

--
Carl Rytterfalk
http://www.pbase.com/rytterfalk
Proud SD9 owner.

 

[Kees Raats]

I would like to have the tool.

I have my camera now for about 10 days and I am still fighting with the color adjustments. If it could be automated for a lot of them it would be a great thing. So please let me (us) know.

Kees

Hello everyone,

I've spent considerable time researching the color characteristics
of the SD9. As a result of this I've made several hue adjustment
profiles that can be used in Photoshop to tweak the colors to a
more accurate value. Not all pictures need color correction though,
it appears to vary with the colors involved, light conditions and
the general exposure level.

I'm planning to construct a web page, demonstrating some effective
Photoshop techniques/workflows of adjusting the colors. This will
include access to my recent hue correction profiles for use with
the SD9, and how to use them in the most effective manner.

How many of you would be interested in such a web page?

I also want to create a small gallery of example images that
includes one or more versions of the images with adjusted colors.
Side by side comparisons will demonstrate the effect of adjusting
the colors. In connection with this I'll need permission from you
SD9 owners out there to use one or more of your sample images as
example images on my web page. I don't own a SD9 myself yet! Those
of you who want to give me a permission to use their images, please
notify me, either by email or by responding to this post. I won't
be using images from users that have not given me a permission. Of
course, the images will only be used on my web page and will not be
used for anything else than pure examples on this specific page.

I have one last request for those SD9 owners who want to help me a
bit further in the study of the SD9 color characteristics. Below is
an test image (1024x768) that I want you to make a monitor
screenshot of. Make sure the room is dark and use average metering.
Set white balance to auto (One can always change this in the RAW
file later, but if auto was used, then at least the auto reading
will be present in the RAW file). Make 4 additional versions with
-2ev, -1ev, +1ev and +2ev. Shoot at smallest resolution (or
medium). I don't need full sized images, it's more important that I
can have the X3F files. The files can be mailed to me. The test
pictures won't be published, they will only be used by me to study
the hue color balance.

I would appreciate any comments and/or suggestions!

Regards,
Geir

Test image:

--
Kees

 

[Dominic Groß]

Hi Geir you made a great job so far and I am sure you will do a good site but but the idea of shooting a color chart directly from the Screen is certainly not a good one. It its not better than letting people make a print of your color chart and then make a picture of that one. There are just to many posibilities if you try to draw a conclusion from the pictures you will get. Most people have bad calibrated Monitors, The Color temprature makes a big difference, also Gammacorrection does. Some Graphicards screw colors up (and a lot of people have those nvidia cards with their damn vga signal quality) and some drivers tend to correct this but therefore manipulate the color. Making pictures of light imiting things like a Screen is always dificult in terms of contrasts, lightening AND blown out colors (this even happens to those guys that make these so called "screeners" in Cinema with real 3 CCD Pro video equipment). Brigther areas are always blown out and higly saturated areas too. Sometimes they don't look like that and you will only notice a small difference to your reference then you maybe make the conclusion that it is the Cameras fault but infact there are some shiftings to because of all the components involved in this process.......

There was another guy here in the Forum who wanted to make a picture of an it8.7 target under Studio Lightening maybe this is a better point to start. And If you have all those pictures don't ignore that you can't fix any problem with the hue/saturation tool. You know that the hue tends to change a bit when lightening changes. But saturation does the same. Best examples for this is a unsaturated birght yellow that changes to a more saturated darker yellow on the picture while nearly all other yellows are correct. And those things are not fixable by changing hues, saturation or Lightening for all colors with the same hue. Those things are very dificult to handle and I have not found a good way in Photoshop so far.

Regards

Dominic
--
http://www.pbase.com/dgross
ICQ UIN: 289647506

 

[Ladehaug]

Gosh, this is fun. More homework.

I am interested and would be willing to pay a bit for the tools. Or
I'll see what I can do to help you get an SD9 reasonably priced.

And I appreciate this. Constructive efforts move things forward.

Laurence

Laurence,

Thanks for the kind words and for doing your homework so fast with the X3F samples - it is greatly appreciated! Selling or charging anyone for my web page with PS tutorials and tools is out of the question! I have my daily professional work from which I earn my salary for a living. My additional work with digital photography is merely a hobby to me. I think it's important that the average amateur photographer can obtain some useful information to make the best of their photos. I won't deny that it would be nice to be sponsored, but more important to me is the help and feedback I can get from current users. I'm going to get my own SD9 anyway, and it's just a matter of time!

Geir

 

[Laurence Matson]

Geir Rune,

Didn't you notice that I just bought one at a real bargain price. You can have the lenses for free. Rick has the specifications for their use somewhere.

Laurence

Gosh, this is fun. More homework.

I am interested and would be willing to pay a bit for the tools. Or
I'll see what I can do to help you get an SD9 reasonably priced.

And I appreciate this. Constructive efforts move things forward.

Laurence

Laurence,

Thanks for the kind words and for doing your homework so fast with
the X3F samples - it is greatly appreciated! Selling or charging
anyone for my web page with PS tutorials and tools is out of the
question! I have my daily professional work from which I earn my
salary for a living. My additional work with digital photography is
merely a hobby to me. I think it's important that the average
amateur photographer can obtain some useful information to make the
best of their photos. I won't deny that it would be nice to be
sponsored, but more important to me is the help and feedback I can
get from current users. I'm going to get my own SD9 anyway, and
it's just a matter of time!

Geir

--
Go Alinghi

--------------------
Current Standings
Alinghi 3
Wind 5
Team New Zealand 0
Kiwi Fans 30

http://www.pbase.com/lmatson/sd9_images

 

[Ladehaug]

You've got mail.

We all need a State of Grace

 

[Ladehaug]

KIM MIN-SU wrote:

 

[Ladehaug]

I'd be interested - I don't have an SD9 myself yet either, but will
probably be buying on in about a month or so...

As almost everything in this world, it requires money. I need to take care of some financing first, and then I'm ready to go for it!

Geir

 

[Ladehaug]

Alright!! Now we talkin'n!

And yes, you can use whatever you want. And I'll try to do this
screenshot test, but I'm not so sure about my screens calibration,

Yeah! But be patient. The page won't be up and running for another week yet!

If you have Photoshop properly installed, then you can use the Adobe Gamma tool in the Windows Control Panel to calibrate your monitor. This is a nice tool. In any case, I can manually set the white balance with the help of the greyscale bar on the left side on the test image. With the help of this greyscale bar, I can even roughly estimate your monitors color temperature by looking at the X3F with a static white balance setting (daylight for example). Unless you have a really bad monitor or you are using some crazy color temperature, then the result will be within reasonable range and usable to me. I will be able to read the approximate hue balance with increasing intensity, which is the main interest for me in this analysis.

Additional tips for shooting the test image screenshot:

To prevent monitor flicker interference, try a higher F stop (use Av mode). This will increase the shutter speed, but noticeably improve the result. So if you have a tripod, then use it! To prevent moire caused by the small phosphoreous R,G and B cells on the monitor, manually adjust focus to be slightly out of focus.

Geir

 

[Ladehaug]

I would like to have the tool.

I have my camera now for about 10 days and I am still fighting with
the color adjustments. If it could be automated for a lot of them
it would be a great thing. So please let me (us) know.

Kees

Just follow the activity in this forum, and you'll soon know!

By the way, have you posted any images for the public yet? I would be happy to see some!

Geir

 

[Ladehaug]

Thanks for your comments Dominic. The main purpose for this test is to study the change in hue versus intensity and how much this varies throughout the hue range. Each of the 16 colors on the test image have the same hue. It would be nice to know the pattern of this when working with the color hue profiles, so that the user is aware of the limitations involved. I'm aware of the differences that can appear based upon different monitor settings, but at least I'll be able to see and account for differences in color temperature by using a fixed white balance in the X3F file. Differences in constrast, brightness and monitor gamma setting is not a problem, as this will only narrow or widen the apparent dynamic range. I can always account for that difference.

I've had the time now to study the X3F files that Laurence was so kind to send me. He sent me 7 different versions, from -3ev to +3ev. None of them looked terrible in my eyes, in fact I was positively surprised by the quality. These samples certainly have helped me a lot to see how the colors respond to the hue adjustment profiles. Now I've got more clear answers to why certain images turn out as they do when I run my hue adjustment profiles on them. So to me (and others as well with the future web page coming up), this is certainly not a waste of time!

Geir

 

[Dominic Groß]

Thanks for your comments Dominic. The main purpose for this test is
to study the change in hue versus intensity and how much this
varies throughout the hue range. Each of the 16 colors on the test
image have the same hue. It would be nice to know the pattern of
this when working with the color hue profiles, so that the user is
aware of the limitations involved. I'm aware of the differences
that can appear based upon different monitor settings, but at least
I'll be able to see and account for differences in color
temperature by using a fixed white balance in the X3F file.
Differences in constrast, brightness and monitor gamma setting is
not a problem, as this will only narrow or widen the apparent
dynamic range. I can always account for that difference.

I've had the time now to study the X3F files that Laurence was so
kind to send me. He sent me 7 different versions, from -3ev to
+3ev. None of them looked terrible in my eyes, in fact I was
positively surprised by the quality. These samples certainly have
helped me a lot to see how the colors respond to the hue adjustment
profiles. Now I've got more clear answers to why certain images
turn out as they do when I run my hue adjustment profiles on them.
So to me (and others as well with the future web page coming up),
this is certainly not a waste of time!

I am still searchin for that tool that was able to do the hue correction in relation to the saturation and lightening... there was one for macintosh (PC Version was announced) by a German company...

Geir

--
http://www.pbase.com/dgross
ICQ UIN: 289647506

 

[Ladehaug]

I am still searchin for that tool that was able to do the hue
correction in relation to the saturation and lightening... there
was one for macintosh (PC Version was announced) by a German
company...

Yes, for a while I was thinking this would be a possibility, but unfortunately I'm afraid it won't work very well in practice.

Tip: Take a decent screen capture of the testimage. Then use the "Select Color Range" in Photoshop. Set a narrow selection range (fuzziness, 10 - 20 will be suitable). Then seek through the image with the color picker and notice how several areas in different color bars are selected. Study the hue, saturation and lightness in the selected areas and you will discover that they are equal. You will also discover that under exposing the image appears to make things better regarding the color separation. Kim put it nicely in one of his posts, by having two options when taking photos with the SD9: 1. Under expose the image and have more accurate colors (better color separation), but with more noise present. 2. Expose the image normally, get cleaner images, but with more inaccurate colors (less color separation).



Testimage captured with the Canon S40. Color separation is quite good, except for the low intensity reds/oranges and parts of the blue colors. The low light reds/oranges is a problem in my S40 images, as orange tones tends to become far too red. This test just confirms the problem with my S40! At least I won't be having that problem with the SD9!

Geir

 

[KIM MIN-SU]

I want to suggest the profile of RAW (not processed) data of X3F or
others. The chromatic property is changed by interpolation process.

Hi Geir you made a great job so far and I am sure you will do a
good site but but the idea of shooting a color chart directly from
the Screen is certainly not a good one. It its not better than
letting people make a print of your color chart and then make a
picture of that one. There are just to many posibilities if you try
to draw a conclusion from the pictures you will get. Most people
have bad calibrated Monitors, The Color temprature makes a big
difference, also Gammacorrection does. Some Graphicards screw
colors up (and a lot of people have those nvidia cards with their
damn vga signal quality) and some drivers tend to correct this but
therefore manipulate the color. Making pictures of light imiting
things like a Screen is always dificult in terms of contrasts,
lightening AND blown out colors (this even happens to those guys
that make these so called "screeners" in Cinema with real 3 CCD Pro
video equipment). Brigther areas are always blown out and higly
saturated areas too. Sometimes they don't look like that and you
will only notice a small difference to your reference then you
maybe make the conclusion that it is the Cameras fault but infact
there are some shiftings to because of all the components involved
in this process.......

There was another guy here in the Forum who wanted to make a
picture of an it8.7 target under Studio Lightening maybe this is a
better point to start. And If you have all those pictures don't
ignore that you can't fix any problem with the hue/saturation tool.
You know that the hue tends to change a bit when lightening
changes. But saturation does the same. Best examples for this is a
unsaturated birght yellow that changes to a more saturated darker
yellow on the picture while nearly all other yellows are correct.
And those things are not fixable by changing hues, saturation or
Lightening for all colors with the same hue. Those things are very
dificult to handle and I have not found a good way in Photoshop so
far.

Regards

Dominic
--
http://www.pbase.com/dgross
ICQ UIN: 289647506

--
SD9 with 24-70DG2.8

 

[Ladehaug]

I want to suggest the profile of RAW (not processed) data of X3F or
others. The chromatic property is changed by interpolation process.

Thanks for the information you sent me. I have noticed that much work has already been done with decoding and analyzing the X3F file format etc. This is of great value and is a very good starting point for further development. Working on the X3F data directly is the optimal way to go, there is no doubt about it. I'll for sure look into this in more detail, and it is very tempting to do some serious programming work in order to create an useful X3F conversion program with emphasis on the color calculations. I started with programming 20 years ago on my first home computers, the Commodore Vic 20 and later the C64 and the Amiga (assembler mostly). 10 years ago I programmed a small optimized graphics library on PC in assembler and C++, with drawing functions that could take both RGB and HSL data as input parameters. But at that time it was all fun and nothing else! Many years later now and programming has been a part of my daily work for some years. Sometimes programming projects have been more stress than fun to work with, therefore I've not been particularly drawn to it in my free time. No matter how you turn it, programming work is rather time consuming, and currently I don't even have the necessary programming tools on my home computer. I need to see what options I have. So give it some time, and I might have something going on with programming!

Geir