Colour calibration process Capture One - x-rite colour checker

[justjay]

Ii looks like "the corresponding ICC profile" was identical for each image, see below ..



Thanks for the two images. Each one has a completely identical Adobe RGB (1998) profile with XYX PCS including identical black and white references and simple, identical, 2.2 gamma tone reference curves.

With identical profiles I would expect the RGB numbers to be different betwixt the two but, for example, the green patch numbers are:

no glass: 163, 239, 109. with glass: 162, 239,108.

Makes no sense to me at all, sorry. Either the RGB numbers should be different or the profiles should be different.

Summat's up or summat!

You have lost me with most of what you are saying. I call them ICC profiles because that's what C1 and xrite calls them..

See here https://imgur.com/a/tLiGfeG

I created the profile via the steps I outlined in my first post.

I also do not know how to use the profile checker thing you mentioned. I used photoshop to look at my 2 images and the green square has very different RGB values.

I managed to upload my ICM profiles here https://easyupload.io/m/rioul1

[Manfred M]

Mostly because I took 2 sample photos of a clients photo album. One with a glass panel to flatten a photo, and one without a glass panel (the photo rolls up slightly) to show her the difference. After I processed them and assigned the corresponding ICC profile, the photos had a obvious colour difference. One was more green and the other was more magenta. I didn't know how to explain the differences to her to I tried to colour match them instead -___-"

I couldn't match them closely so I gave up and processed the whole album without colour edits apart from the ICC profile. She didnt ask questions.

Let's not get into ICC profiles or anything that complicated, you have a completely different issue that you need to address.

First of all, basic colour theory; magenta and green are complementary colours. If you have a magenta colour cast, you need to add green (which is the same thing as subtracting magenta) to get a neutral balance. The same goes for when you have a green colour cast, you need to add magenta (which is the same thing as subtracting green).

Shoot the gray card under the glass and without the glass and pull a white balance off both. Check the Kelvin temperature and the tint values and use those as appropriate. Every other studio photographer I have ever worked with does this.

[dje]

Obviously, but why confuse less knowledgeable folks by mixing the two together in a single sentence?



Adjustments where and to what? I'm thinking not ...



No orthogonal blue-yellow and magenta-green axes to be seen here either:

https://www.handprint.com/HP/WCL/labwheel.html
.

Ted the lines crossing the Planckian locus on your u'v' diagram are lines of constant Correlated Color temperature. On a uv plot, they are orthogonal to the Planckian locus as far as I recall. The Tint value is a measure of how far off the locus the actual illuminant color lies along a line of constant CCT. It an be positive or negative depending on which side of the locus it lies. This relationship forms the basis of calculations used in editing software to convert between Illuminant color (white point) and Temp/Tint. There is no standard unit for Tint though.

Dave

[dje]

Manfred, it sounds like we are taking at cross-purposes. By "camera profiles" were you talking about Adobe DCP files of type .dcp. Reason I ask is that can not see how two identical profiles of type .icc/.icm can correct two difference color casts.

As we should know, ICC profiles have an input side and a output side. While it is reasonable that the output sides would be the same for given output device, it is much less reasonable that the input sides would be the same for two different color casts or, in your case, two widely-differing Kelvin lighting values ...

Now that Jay has posted the two profiles, you can open them up with ICCProfileInspector. CaptureOne uses camera calibration profiles in the icc format but they serve the same purpose as Adobe DCP profiles. You will notice that they use A2B0 type LUT's rather than matrices. The LUT's do seem to be slightly different in the two profiles.

Dave

[dje]

Shoot the gray card under the glass and without the glass and pull a white balance off both. Check the Kelvin temperature and the tint values and use those as appropriate. Every other studio photographer I have ever worked with does this.

That's what I said in posts 3 and 6 Manfred

[xpatUSA]

Ted the lines crossing the Planckian locus on your u'v' diagram are lines of constant Correlated Color temperature.

Yes, as is well-known.

On a uv plot, they are orthogonal to the Planckian locus as far as I recall.

I'll take your word for it but there does seem to be some variance in slope here:

https://cran.r-project.org/web/packa.../isotherms.pdf see page 2.

The Tint value is a measure of how far off the locus the actual illuminant color lies along a line of constant CCT.

yes, as is also well-known.

It can be positive or negative depending on which side of the locus it lies. This relationship forms the basis of calculations used in editing software to convert between Illuminant color (white point) and Temp/Tint. There is no standard unit for Tint, though.

Dave

I've seen "mired" used in that context, but I'm not sure that it's usage is right.

See:

https://en.wikipedia.org/wiki/Mired
.

[Manfred M]

That's what I said in posts 3 and 6 Manfred

Just re-enforcing the same opinion.

Sometimes we get too far down into the rabbit hole to see what the actual question is. I think that is what has happened this time diving too deeply into theory.

[xpatUSA]

I've seen "mired" used in that context [tint], but I'm not sure that it's usage is right.

Good ole' Doug Kerr stated a while back:

"As I mentioned, we mostly see the unit mired in connection with chromaticity shift (in a color temperature perspective). It is only fully apt for a shift of chromaticity in the direction (on a chromaticity chart) along, or parallel to, the Planckian locus.

Sometimes, however, we find changes in the perpendicular direction (along a line perpendicular to the Planckian locus in the region where we are working), or in some other direction, denominated in the unit mired. That usually is taken to mean a unit of distance (on the CIE u-v chromaticity chart) equal to the distance (in that region) corresponding to a color temperature (or correlated color temperature) change of one mired. That usage is deprecated."

I also vaguely remember it being discussed in the context of ACR's "tint" adjustment units long ago .. sorry, no reference.

[xpatUSA]

You have lost me with most of what you are saying. I call them ICC profiles because that's what C1 and xrite calls them..

Yes, I also call them "ICC profiles" - files which end in either .icc or .icm.

I created the profile via the steps I outlined in my first post.

I also do not know how to use the profile checker thing you mentioned. I used photoshop to look at my 2 images ...

Which two images are you referring to? The two I downloaded have identical green squares.

... and the green square has very different RGB values.

Then we must be talking about different images and by now I how idea which are what, sorry.

I managed to upload my ICM profiles here https://easyupload.io/m/rioul1

Yes, Dave and I looked at those. they are different to the ones embedded in the first two images you posted. All quite confusing.

[dje]

Yes, I also call them "ICC profiles" - files which end in either .icc or .icm.



Which two images are you referring to? The two I downloaded have identical green squares.



Then we must be talking about different images and by now I how idea which are what, sorry.



Yes, Dave and I looked at those. they are different to the ones embedded in the first two images you posted. All quite confusing.

Ted, camera calibration profiles (whether they be dcp or icc/icm) are not normally embedded in exported jpegs. The images that Jay posted both contained the standard Adobe RGB profile which, as I'm sure you know, is used by color managed viewing software to accurately reproduce the appearance of the image as it was after final editing. The camera calibration profile's job is done after raw conversion has been completed.

Dave

[dje]

Good ole' Doug Kerr stated a while back:

"As I mentioned, we mostly see the unit mired in connection with chromaticity shift (in a color temperature perspective). It is only fully apt for a shift of chromaticity in the direction (on a chromaticity chart) along, or parallel to, the Planckian locus.

Sometimes, however, we find changes in the perpendicular direction (along a line perpendicular to the Planckian locus in the region where we are working), or in some other direction, denominated in the unit mired. That usually is taken to mean a unit of distance (on the CIE u-v chromaticity chart) equal to the distance (in that region) corresponding to a color temperature (or correlated color temperature) change of one mired. That usage is deprecated."

I also vaguely remember it being discussed in the context of ACR's "tint" adjustment units long ago .. sorry, no reference.

I don't think it's normally used in relation to Tint. It is specifically defined in relation to deg K, ie 1000000/deg K. Adobe specifiy the use of mired Kelvin values (in their dng spec) with their dual illuminant dcp profiles. They are used for the purpose of interpolating between the matrix values for the two different illuminant white points.
Dave

[xpatUSA]

Ted, camera calibration profiles (whether they be dcp or icc/icm) are not normally embedded in exported jpegs.

Thank you for your note Dave. However, each of the apps that I commonly use (with it's default settings) does embed an ICC profile into an exported JPEG. So now, confused as ever, what is meant by "normally"?

The images that Jay posted both contained the standard Adobe RGB profile which, as I'm sure you know, is used by color managed viewing software to accurately reproduce the appearance of the image as it was after final editing.

Ok, I'll go back and look at the green patch in those two images again - which I said was the same and Jay said was different. I had them open as layers in the GIMP. I'll open them both in a different app this time.

Hoping that we can get to the bottom of this rabbit hole in spite of Manfred's warning in his #28.

The camera calibration profile's job is done after raw conversion has been completed.

Dave

[xpatUSA]

I don't think it's normally used in relation to Tint.

Which is why I emphasized "That usage is deprecated" ...

[xpatUSA]

Ok, I'll go back and look at the green patch in those two images again - which I said was the same and Jay said was different. I had them open as layers in the GIMP. I'll open them both in a different app this time.

Hoping that we can get to the bottom of this rabbit hole - in spite of Manfred's warning in his #28 ...

... and here we are ... My mistake to claim not-different green patch RGB values.

Opening the two images in RawTherapee, the green patches have different values thereby agreeing with Jay's photoshop experience. Also, when opened as separate images (not layers) in the GIMP, the green patch values are different.

Apologies to all for not realizing how CMS works/ doesn't work in the GIMP layer system ...

[dje]

Thank you for your note Dave. However, each of the apps that I commonly use (with it's default settings) does embed an ICC profile into an exported JPEG. So now, confused as ever, what is meant by "normally"?

Ted I was making the distinction between camera calibration profiles (which the icc calls input profiles) and standard color space profiles such as sRGB, Adobe RGB, ProPhoto RGB etc. The input profiles are not normally embedded in an exported jpeg whereas the others commonly are.

Dave

[xpatUSA]

Originally Posted by dje Ted, camera calibration profiles (whether they be dcp or icc/icm) are not normally embedded in exported jpegs.

Ted I was making the distinction between camera calibration profiles (which the icc calls input profiles) and standard color space profiles such as sRGB, Adobe RGB, ProPhoto RGB etc. The input profiles are not normally embedded in an exported jpeg whereas the others commonly are.

Dave

Ok, Dave. If we're going to stick with SPV (Standard Photographic Vague) where "input profile" means anything, then:

In the quote you said that "dcp or icc/icm" are not normally embedded in exported jpegs. I can only take that as meaning that neither dcp, icc, nor icm are normally embedded.

Now you are saying that "the others", which I can only take to be icc/icm, normally are.

[dje]

Ok, Dave. If we're going to stick with SPV (Standard Photographic Vague) where "input profile" means anything, then:

In the quote you said that "dcp or icc/icm" are not normally embedded in exported jpegs. I can only take that as meaning that neither dcp, icc, nor icm are normally embedded.

Now you are saying that "the others", which I can only take to be icc/icm, normally are.

I think I'm done, sorry ...

Ted what I actually said was "camera calibration profiles (whether they be dcp or icc/icm) are not normally embedded in exported jpegs". The operative words here are "camera calibration profiles". This covers dcp profiles in the case of Adobe and icc/icm profiles specically used for camera calibration as with Capture One. The International Color Consortium defines an input profile as per this link. These are the icc profiles that are not normally embedded in an exported jpeg. icc profiles for sRGB, Adobe RGB etc are.

Dave

[justjay]

Let's not get into ICC profiles or anything that complicated, you have a completely different issue that you need to address.

First of all, basic colour theory; magenta and green are complementary colours. If you have a magenta colour cast, you need to add green (which is the same thing as subtracting magenta) to get a neutral balance. The same goes for when you have a green colour cast, you need to add magenta (which is the same thing as subtracting green).

Shoot the gray card under the glass and without the glass and pull a white balance off both. Check the Kelvin temperature and the tint values and use those as appropriate. Every other studio photographer I have ever worked with does this.

**edited

Is there a reason why I cannot pull the WB from one of the squares of the color checker? Just trying to understand. Is it because none of the grey squares are exactly 18% grey?

Based on my understanding of what Dave has written, the process for using the grey card together with the colour checker is as follows:

1. Capture grey card & color checker card
2. Set WB on grey card and copy to color checker card
3. Change ICC profile of color checker card to "No colour correct" and change curve to "Linear Response"
4. Export as tiff ensuring camera profile is selected
5. Use colorchecker camera calibration software to create ICC-TIFF profile
6. Restart Capture one and apply ICC profile to image
7. Copy WB settings from grey card and apply to images

I'm clarifying because I feel like I am copying WB twice. Once for colour card and once when I apply to image. And when I roughly followed this process, my image ended up very warm...

[justjay]

Thank you to all for your input too. Some of the discussion is way over my head but I appreciate it regardless. It is something for me to read up on in future.

[Manfred M]

**edited

Is there a reason why I cannot pull the WB from one of the squares of the color checker? Just trying to understand. Is it because none of the grey squares are exactly 18% grey?

Based on my understanding of what Dave has written, the process for using the grey card together with the colour checker is as follows:

1. Capture grey card & color checker card
2. Set WB on grey card and copy to color checker card
3. Change ICC profile of color checker card to "No colour correct" and change curve to "Linear Response"
4. Export as tiff ensuring camera profile is selected
5. Use colorchecker camera calibration software to create ICC-TIFF profile
6. Restart Capture one and apply ICC profile to image
7. Copy WB settings from grey card and apply to images

I'm clarifying because I feel like I am copying WB twice. Once for colour card and once when I apply to image. And when I roughly followed this process, my image ended up very warm...

Not nearly that complicated.

1. Take a shot if your subject with / without the glass. For the glass shot, it must be under the glass.

2. In C1 pull a white balance off of one of the neutral squares; I tend to favour the middle to lighter tones simply because there is more data there. I often just use the gray card. As long as I do not change my lighting setup, I will continue to use that reading.

In ACR I use the built-in automation to copy the WB settings to all of the images (one with and one without the glass). I'm pretty sure that C1 will have similar functionality. Consistency is great that way.

I use the custom profile I created with the x-Rite software for Adobe Camera Raw, as that is my usual convertor. I have only even made one profile and reuse it all the time. The one time I made one for C1, I followed very much the same process.