How to get a super result using Film Negative?

[jdc (Desmis)]

First excuse my bad english, White balance is a complex thing, and for me write in english is difficult

This method by optimizing the variance of the R and B channels is one of the many methods I have experimented with.

We find it in the university literature.

I was very interested in her, but after reading the essays that I could see, it appeared that she did not solve the problem of images with a bad “green”. Which does not mean that we can not use it. especially in this specific case “Film Negative”

The second problem is that of relevance : because the variance (or standard deviation) is the minimum that the result is always good?

On this basis I did other research in academic documents and found a study without algorithm or code.
The idea seems close to that evoked but goes further.
Instead of RGB channels we use xyY, which is more relevant in terms of colorimetry,
And instead “variance”, we use a comparison of samples on the one hand within the image, on the other hand from defined spectral colors, this comparison is realized dy a “Student” test.

In the case of “autowb” (whose algorithm has not changed since 1 year and I always wait to merge with dev), I compare a sufficient number of samples from more than 150 areas on the image and 200 reference colors
This comparison is relized by changing “Temp” : this variations make a change in xyY values of image, and in xyY values of spectral datas . This algorithm is complex and needs a lot (200) of spectral data in the visible domain
The best result is for Student minimum.

But, because there is a big “But”, these algorithms consider that the green is good or that it is necessary to make a manual adjustment … what is the height for a WB auto

I search a long time and finally create 2 loops that interact with each other, one for the green (Y), and one for Red and Blue (x y).

In more of 95% case, result is very good, but in some cases where Illuminant is not with a good CRI (color renderind index) - some LED, some Halogen…, algorithm fails.

It must be remembered that the problem of the WB is mathematically indeterminate which explains : a) the number of works on this subjects; b) the imperfection of the results

jacques

[obe (Olaf Berthelsen)]

@rom9

Now that you make me think of it, maybe i should automate the last step, so that when picking the gray spots, the filmnegative tool also implicitly triggers a spot-WB on one of those spots …

That is a very good idea.
In this way, the user doesn’t risk to forget to apply the white balance tool again. The need to reapply seems strange unless the user has read your thorough explanation on the subject……

Before posting my question to you, I tried to pick the white/black spots and then reset the white balance tool or skip resetting the white balance tool. There is clearly a difference, which I didn’t understand then. However, it can be difficult to decide which version is the best because the image needs a lot of further editing.

Thank you for the explanation and all your effort. I think that you and the other RT developers do a fantastic job.

[Morgan_Hardwood]

@rom9 docs updated.

I’m not very fond of calling them “white and black spots” since that’s not what they are. How about renaming the button to “Pick neutral spots” and leaving the explanation to the docs?

[rom9]

Agreed, much better. I’ll change it this evening.
And what about the button tooltip? It currently says:
Calculate exponents by picking two neutral reference spots in the image; one white (light gray) and one black (dark gray). The order does not matter. The exponents will be updated after the second spot is picked.

should I remove the “white” and “black” terms from there, too?

[Morgan_Hardwood]

@rom9 I’ll do it if you don’t mind, as it also requires changing all translations.

[rom9]

@cuniek

Silver does not “mirror” but appears as brown (blue after inversion) sediment. I can deliver examples if needed.

I didn’t know about this problem, thanks for the explanation. I’m afraid that color shift is quite “spotty” and not constant across the frame, so it would be very hard to solve by software means.
Just a wild idea: since we’re talking about B&W negatives i suppose, what about shining an orange/yellow light through it, instead of white? If the problem is more evident in the blue channel, removing it from the source backlight could mitigate that.

@jdc

In more of 95% case, result is very good, but in some cases where Illuminant is not with a good CRI (color renderind index) - some LED, some Halogen…, algorithm fails.

Hi jdc, thanks for your explanation. I know nothing about color spaces, so i’m talking as a complete ignorant here, but that is the exact use case that makes me dubious about auto-wb methods. I have some concert pictures that were shot with an awful red or purple LED light. The pictures have a strong tint all across the frame. How can an autowb method restore the original tint? I would rather take the color balance from another “reference” frame from the same film, and apply that to the other frames. (i’m working on it at the moment and i’ll post some examples later).
Anyway i’ll try the hue variance method as soon as i can get my hands on the gmic filter

[rom9]

@Morgan_Hardwood

I’ll do it if you don’t mind

Sure, thanks!

[afre (Alan)]

@rom9 @Iain @David_Tschumperle

@Iain

why Auto WB by hue variance is not showing up in Testing → Iain Fergusson

@afre

Same here. There was an update bug not long ago…

FYI, still unavailable.

[David_Tschumperle (G'MIC staff)]

It should be OK now. I’ve been forced to update ‘manually’. I still don’t know why the automatic update didn’t work.

[afre (Alan)]

Is it related to the earlier bug where my new filters didn’t appear for days? Or did you fix that one bug but it turns out it didn’t solve everything? Anyway, I can’t wait to try @Iain’s filter.

[rom9]

Thanks, now the filter shows up in the list and everything works fine

@jdc : i made some tests with the hue variance method, and they seem to confirm our doubts with pictures that have a very un-balanced illumination.

This one came out pretty good:

… and also this one (just needs some contrast):

… but this one looks a bit off:

In reality, this scene was lit by a very strong red light from the right, and an orange/gold light from the front, so everything had a much stronger red tint.
This is the problem i think: how can an auto-wb method (however good it may be) figure out all this stuff, without any hint about the original scene?

So, this is what i do instead. First, i take a reference picture, pick the 2 neutral grey spots to calculate negative exponents, and manually set white balance via spot-wb:

Then, i blindly paste the same film negative and white balance parameters on all the other pictures from the film roll, and get this (note the exact same WB numbers on the right):

… which looks pretty accurate. You can get a hint in the upper left corner:

it’s an “emergency exit” lamp (with huge amounts of coma from my lens at f1.4 ), one of those green/white signs, i think they’re quite standard across europe at least.
To me it looks reasonably good, despite the rest of the image having very un-natural colors.

I know this is a super-dumb method, there’s nothing automatic, you have to manually setup the reference picture (it must not be a macbeth chart, any picture with 2 neutral gray spots will do), but afterwards you can set-and-forget all the other pictures in the film roll and get reasonable results…

[Iain]

Thanks for giving my filter a test.

It occurs to me that calling it Auto White Balance is not really accurate, even though that was the intent. It colour enhancement that happens to get close to a white balanced image in many cases.

It is trying to maximise the number of hues in an image, so when an image naturally has very few hues it is going to produce unpleasant results.

However, in some cases, it produces nice results that probably did not look like the real scene. For example, this image from DPReview’s Nikon D5 review camera WB is probably most accurate but I prefer the hue variance version.

Edit: Also, with a standard whit balance picker you need pick a neutral grey, but with this method you need an area with a range of hues. I’ve found that peoples faces often have enough variation in hues to produce good results. So you ‘white balance’ without any neutral grey.

[afre (Alan)]

This is getting a bit off-topic won’t you say, since this thread is about film negatives. I would love to see a new thread on the topic of auto wb, starting with a comparison between @Iain’s and @jdc’s methods. Personally, I believe they have different intents.

[Iain]

Move white balance discussion here please:

White Balance method discussion

Starting this thread so the thread about Processing Film negatives is not contaminated. Good suggestion @afre See here: How to get a super result using Film Negative? - #44 by Iain My Hue Variance method described here: How to get a super result using Film Negative? - #49 by Iain

[Morgan_Hardwood]

RawTherapee and RawPedia updated - “white and black” changed to “neutral”.

[troodon (srowed)]

DSC_1390.jpg.out.pp3 (11.3 KB)

Here’s a version with just playing with the settings, without using Film Negative.

[troodon (srowed)]

corrected the colour a bit more here.

Ignore these as I messed up. I’ll post a corrected version.

[troodon (srowed)]

This will hopefully solve the problems with the other two versions I posted.

[obe (Olaf Berthelsen)]

Hi’ @troodon
Thank you for your post.
My first reaction: This is a surprisingly good result! Why do you need film negative when you can create such a good result just using the basic RT tools? However, after a closer look at your pp3 it seems to me that it is not so straightforward to get the result using only the basic tools.
First of all you apply a reverse tone curve:

And you also use several other RT tools including RGB curves (R and G channel). Therefore, I think that it requires much less RT expertise to obtain a good result using film negative.
@rom9 what do you think?

[troodon (srowed)]

One issue with the raw file is that the color balance does not seem consistent from left to right, being a bit warmer on the left than the right of the frame. Whenever I’m digitizing film, whether positive or negative, I always shoot a “blank” of just the light source, which in my case is a high-quality 5600K LED light panel. I position the film, then remove it to ensure that the area I’m shooting matches with the film. This “blank” does two things - first it’s a white balance check in RawTherapee, which I use for the film (at least if it’s color transparency film). Second, it provides a reference to ensure that the brightness and color are consistent across the frame. This can be done easily by referencing the histogram on the camera LCD. If it’s a very narrow histogram, it means that the brightness is consistent in all parts of the frame. If it’s wider, it means there’s more variation and you need to make corrections. This variation can be caused by the light source being brighter on one side of the frame, by specs of dirt on the light table, or even by using a wide aperture that has darkening on the edges of the frame. With my Nikkor 60 f2.8 micro lens I need to stop down at least to f5.6 to eliminate this vignetting, and I generally use f8 - f11. This screenshot of a “blank” shows that the exposure and color are consistent through the frame.