White balance of digitalized color negatives

[Einst_Stein]

Advertisement (gone after registration)

I saw suggestions to use the white balance drop to click on the negative orange border, and implies that is the white balance.

Correct me, but I think there are two issues in white balance. The first is the color negative orange base (border) color, which has no clue of the lighting condition during the exposure.  The second is the light condition during the exposure.

Assuming the negative has a grey card reference in the picture, another way to do the white balance is to click the white balance drop on the grey card.  But this method may still have a problem. I assume the grey card method was handled by de-coloring the whole scene assuming the orange cast was proportional to the brightness of each pixel, not a uniform constant across the whole scene.

In another word, this is a two variables problem.

How is it handled by the scanner software? How professional film scanning experts handle it?


 

[frame-it]

9 minutes ago, Einst_Stein said:

 

How is it handled by the scanner software? How professional film scanning experts handle it?


 

e.g.

https://www.silverfast.com/show/negafixguide/en.html

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[wda]

I would have thought it would be better to correct white balance after scanning.

[Exodies]

Depends if you want a technical result or you want it to look good. For the first, the scanner software might have a better idea of what’s going on.

[wda]

9 minutes ago, Exodies said:

Depends if you want a technical result or you want it to look good. For the first, the scanner software might have a better idea of what’s going on.

But if you are merely sampling a calibrated grey card, it shouldn't make any difference  when correcting white balance. It is easy enough to test and prove. Request OP posts his comparative results.

Edited March 27, 2020 by wda

[250swb]

Use ColorPerfect as a plugin after scanning, it is designed to compensate for the film base. The price saves hours of angst.

[philipus]

Advertisement (gone after registration)

Hi Einst

Are you trying to achieve good colour of a digitised image? If so, use the eyedropper tool (I use Adobe Camera Raw which works well) to find a place where the three RGB values are the same or very similar and click there. That will (usually) give a good starting point for further colour correction.

Or are you wondering about the feature in for instance Vuescan to "lock film base color" (and here)? I used to do this but found that it is a waste of time because it requires extra previews and because I use ColorPerfect for inverting the image which gives considerably better colour than Vuescan does.

br
Philip

 

On 3/27/2020 at 8:44 AM, Einst_Stein said:

I saw suggestions to use the white balance drop to click on the negative orange border, and implies that is the white balance.

Correct me, but I think there are two issues in white balance. The first is the color negative orange base (border) color, which has no clue of the lighting condition during the exposure.  The second is the light condition during the exposure.

Assuming the negative has a grey card reference in the picture, another way to do the white balance is to click the white balance drop on the grey card.  But this method may still have a problem. I assume the grey card method was handled by de-coloring the whole scene assuming the orange cast was proportional to the brightness of each pixel, not a uniform constant across the whole scene.

In another word, this is a two variables problem.

How is it handled by the scanner software? How professional film scanning experts handle it?


 

[tommonego@gmail.com]

The nice thing about ACR is you can save your base color settings, so using what I did as experiments recently, I had some recent pictures on Portra 160 and some older images on Agfa Optima 100, so I used the eye dropper for color temp, inverted the curve and I was very close on the Agfa film with a light base, and reasonably close on Portra which has a dark base, required a little more work, but I saved the settings. But using second negatives from the same films, the saved XMP files brought the images very close, so the saved XMP files were repeatable saving time. 

First is Portra 160 second Agfa Optima 100

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

[jolsen]

Einst_Stein I believe you are right about the 2 variables. There may even be 3, the 3rd being the exposure setting when you scan. But this one you may be able to easily control.

This is what I did to get around the first problem with the orange mask.

I happened to have a negative shot of the Kodak Q13 Grey scale on the negative film I used the most. I also had a shot on slide film of the same grey scale. They were both shot in sunlight with correct exposure. I scanned both using auto exposure on my Nikon 5000 ED. Both were scanned as positives, so the negative came our with orange mask and everything. 

In Lightroom, I sampled the A, M, and B patches of the slide film, and adjusted the curve or the negative to obtain the same values for the same patches. I adjusted the color curves one by one, and it was a lot of back and forth. It worked surprisingly well. The slide and the negative, after applying my curves, looks very similar.

When I scan negatives of the same film type, shot in sunshine, the results are very good. Colors are (almost) perfect, no matter there are any white in the scene or not. NikonScan only managed to get this good colors if I was shooting snow.

Now when scanning negatives which were not shot in the sun, the results are not as perfect. I believe this has to do with your 2nd variable. I believe the way around this is to shoot the grey scale in each lightning condition, and then make a curve for each. This may of course not be possible, if you don't have that shot already, or you are not able to get the same film type now. I guess, theoretically, it should be possible to just adjust the color temperature after applying the sunshine curve, but I haven't tried this yet.

I am now scanning all my old negatives. I am only scanning negatives of the same film type as my reference scan. The exposure of the scanner is set to the same value as when I scanned the reference scan, which means the only variable I have left now is the lightning condition.

You may find my KODAK 400UC preset for Lightroom in another post.