Copying color negatives with camera

I figured I would go ahead and post this as an experiment in progress. My basic setup is a Fuji X-t1 equipped with a shutter release cable, Minolta 50mm macro lens, 1:1 extension tube and a Metabones adapter that has a lens mount(important to have it because it centers the lens)

From there, I have a Minolta brand set of bellows, focus rail, and slide copier attachment. The bellows does not have much use for me because I ideally prefer to have a 1:1.5 ratio, and I don't have a lens that could provide me with one when attached to the bellows. So instead I use only the focus rail with slide copier attached. Because my slide copier has a bellows that attaches to the lens I don't have to work in a darkroom.

Beyond the camera, and focus rail, there is a light-box that supplies a 5000k light source, and a off camera flash hooked up to the PC terminal on my camera(I'm going to change it to a TTL cord later because the cord I am using has been showing signs of corrosion on the contacts). The flash is pointed at the light box so that it bounces off the box and on the back of the slide copier attachment.

I'll add some pictures later to show the setup.

When I was copying slides, it was easy to do, no problems with color correction, but now moving on to copying color negatives is a whole new ball game. So after weeks of reading, agonizing over how to do it properly I have come up with some theories to test.

Most solutions have suggested to shoot an undeveloped portion of the film and apply an auto white balance to it in GIMP, then apply the same settings to the developed film. This does not work at all. Once I invert the negative to a positive, there is an ugly green tint. Trying to white balance the "positive" image for the green gives me some bad coloration.

I tried custom white balancing in camera, not helpful. I tried white balancing in camera to an undeveloped negative, didn't work either. Then I tried setting the white balance to the lowest Kelvin possible. Now I have made some progress.

Seeing how setting my camera's white balance to the lowest possible Kelvin could compensate for the orange, I decided to try something else use a blue color correction filter. I set the white balance to 5000k so it matches the light-box, and then I attached a 80a color correction filter to the macro lens. I got useable results that could be adjusted in GIMP using the levels and autowhite balancing tools.

From what I have read about the negative to positive process, it sounds like a set of color correction filters are used in creating a positive print.

If there are any expert on here who can chime in, from what I read, it sounds like the best way to use color correction filters is to have the filters alter the light that hits the back of the film negative? What about having the filters on the lens which filters light that has already passed through the negative? What would be the next step, buy a set of color correction gels for the flash unit? Or would a set of lens filters be adequate for my purpose?

I look forward to seeing the photos of your set up and also to hearing about how you've overcome the current technical difficulties.

Thanks for taking the time to post this.  I've been looking into doing this myself, and had planned nearly the same approach.  My plan is to make a platform to hold the camera (Sony a6000) and lens (telephoto zoom with macro adapter -- at ~6" I can fill the frame).  On the opposite side will be a flash (Yongnuo 560 III, operated by rf remote) shining through a diffuser plate where the negative will be mounted.

Using a zoom lens means I don't have to build in any rails, and I can easily turn this platform into a box with lid if I get tired of turning out the lights.

I'm especially interested in (troubled by) your experience with the white balance, though.  I also planned to try the camera calibration of white point using undeveloped film approach, at least as one option.  Otherwise, I figured I'd shoot in RAW, and figure out how to tweak WB in post.  But I'm not big into PP, and don't have any decent software (LR, etc.).

I do use GIMP, though, but only for minor tweaks on jpegs.  For tweaking WB on a RAW photo, I'd used Sony's free software.  Maybe GIMP doesn't have the latitude needed to make such a large adjustment to WB(?).

My understanding of WB adjustment is that color temperature is only a part of it.  Color temperature just describes a line on the CIE color chart that corresponds to black body radiation.  But, if you're shining your light through negative film stock (essentially a color filter), you're departing from this line.  On the Sony a6000, you can adjust white balance by color temperature, but then you can also make color corrections beyond that, which I'm assuming I'd need to do.  Is this an option on your camera?  It still seems to me like the proper WB adjustment will do the trick, but until I try it, I won't know the limitations.

I'll be very interested to hear of your progress, and thanks again for posting!

TJ61 wrote:

Thanks for taking the time to post this. I've been looking into doing this myself, and had planned nearly the same approach. My plan is to make a platform to hold the camera (Sony a6000) and lens (telephoto zoom with macro adapter -- at ~6" I can fill the frame). On the opposite side will be a flash (Yongnuo 560 III, operated by rf remote) shining through a diffuser plate where the negative will be mounted.

Using a zoom lens means I don't have to build in any rails, and I can easily turn this platform into a box with lid if I get tired of turning out the lights.

I'm especially interested in (troubled by) your experience with the white balance, though. I also planned to try the camera calibration of white point using undeveloped film approach, at least as one option. Otherwise, I figured I'd shoot in RAW, and figure out how to tweak WB in post. But I'm not big into PP, and don't have any decent software (LR, etc.).

I do use GIMP, though, but only for minor tweaks on jpegs. For tweaking WB on a RAW photo, I'd used Sony's free software. Maybe GIMP doesn't have the latitude needed to make such a large adjustment to WB(?).

My understanding of WB adjustment is that color temperature is only a part of it. Color temperature just describes a line on the CIE color chart that corresponds to black body radiation. But, if you're shining your light through negative film stock (essentially a color filter), you're departing from this line. On the Sony a6000, you can adjust white balance by color temperature, but then you can also make color corrections beyond that, which I'm assuming I'd need to do. Is this an option on your camera? It still seems to me like the proper WB adjustment will do the trick, but until I try it, I won't know the limitations.

I'll be very interested to hear of your progress, and thanks again for posting!

On my X-T1 I can set the WB to a custom color temperature, So the higher Kelvin's have a reddish color to balance out "blueish" lighting and the lower Kelvins use a blueish color to balance out the orange/reddish light.

Also for any custom WB temperature I pick on my camera I can fine tune the setting such that I can move the WB towards cyan,blue,red or yellow depending on my preferences.

My software right now is the Raw File Converter that came packaged with my camera, it has a useful histogram that tells me the balance of red,blue and green.

I have had some limited success using the lowest possible kelvin setting and pushing the blue tint up to the maximum on the WB setting. But there are some problem negatives that occasionally show up.

Attached is a picture of my current set up

Assuming you have Photoshop (an old version will do).

First check that your image is 16 bits per channel. If not, change it in Image Mode.

Invert.

Go to Curves. Under Options, enable "Snap Neutral Midtones". Click OK. This will probably give a rather pale image with nearly correct colours.

Go to Levels. Use the middle slider to darken the image if necessary. Then you can adjust the colour balance to taste by using the middle slider for red, green or blue separately. Very likely a slight touch of red is all you need.

D Cox wrote:

Assuming you have Photoshop (an old version will do).

First check that your image is 16 bits per channel. If not, change it in Image Mode.

Invert.

Go to Curves. Under Options, enable "Snap Neutral Midtones". Click OK. This will probably give a rather pale image with nearly correct colours.

Go to Levels. Use the middle slider to darken the image if necessary. Then you can adjust the colour balance to taste by using the middle slider for red, green or blue separately. Very likely a slight touch of red is all you need.

That method is producing interesting results.

I took a couple of shots of the same negative with different white balance adjustments in-camera. It looks like on the histogram that the balance of red blue and green are at different "phases" horizontally they don't line up.

After a couple of shots with different settings I went with the one with a histogram that looked the most centered of all three colors. The screenshot doesn't say but its the second picture from the right.

Here is the result.

I noticed some yellowing on some of my pictures, what is going on?

AdHoc007 wrote:

I noticed some yellowing on some of my pictures, what is going on?

That looks like poor processing. Probably the bleach-fix has not completely removed the silver.

You could try copying the films first, then putting a test strip in fixer. Wash thoroughly, final rinse in distilled water. The blotches may clear. Try copying again to see if there is any improvement.

D Cox wrote:

AdHoc007 wrote:

I noticed some yellowing on some of my pictures, what is going on?

That looks like poor processing. Probably the bleach-fix has not completely removed the silver.

You could try copying the films first, then putting a test strip in fixer. Wash thoroughly, final rinse in distilled water. The blotches may clear. Try copying again to see if there is any improvement.

I don't have a photo lab, but I suppose I could go down to the local photo lab and see if they have a specialist who can do it. I wouldn't normally go out of my way to try and fix this but this photo is on a film strip with another photo that I do actually want to preserve.

zirtico wrote:

I've done this before. I find that even after colour correction, you may still need to adjust the curves and levels, since the final image is always a little low contrast.

My first step is to shoot the orange mask of the negative first as WHITE in the custom white balance. This lets the camera do as much of the mask removal as possible. Then you shoot your frames on that particular brand of film (since the mask is unique to the brand). Then the moment you pull your RAW files into your workflow and invert the image, the filmstrip around the frame should be black (since it was white to begin with), and the image should at least resemble the original.

Then you can adjust saturation/contrast to taste. This is the most consistent method I've found.

This was the general method I was hoping to use, and tried it for the first time over the weekend. I think it's a good starting point. It took me a little while to convince myself what part of the negative I should use to do my white balancing. I kept thinking I should use an area that would come out white in the print, but finally concluded that I need to use an area that would come out black (i.e. area in between frames).

Here's the evolution...

RAW import, and using camera's (wrong) WB setting.

Reset WB using area to the left, outside of frame, then saved as jpeg

After inverting colors of the jpeg in GIMP -- kind of washed out still

Increased contrast first, then used Curves to increase brightness, then added a little saturation (that's Lake Louise down in the valley, btw, with the Chateau on the far side)

This photo doesn't quite look like the print (it actually looks better), and I don't know how "real" it looks, but I definitely have enough "proof of concept" to start building a platform to hold the pieces.

One thing I need to figure out is how best to tone down my flash. Right now, it's about 2" behind the negative, and at it's lowest power I had to use f/25(!) to get the proper exposure. I guess the first step is just to increase the distance, though I was hoping to keep things compact.  I'd like to use f/11 - f/16 to give enough DOF.

TJ61 wrote:

This photo doesn't quite look like the print (it actually looks better), and I don't know how "real" it looks, but I definitely have enough "proof of concept" to start building a platform to hold the pieces.

Nice job!

One thing I need to figure out is how best to tone down my flash. Right now, it's about 2" behind the negative, and at it's lowest power I had to use f/25(!) to get the proper exposure. I guess the first step is just to increase the distance, though I was hoping to keep things compact. I'd like to use f/11 - f/16 to give enough DOF.

Try placing one or more sheets of white paper in front of your flash.

3) By a long long way the best results will be obtained by pre-filtering the lightsource in such a way as to get a capture that matches the native colour characteristics of the cameras sensor and accounts for the colour temperature of the lightsource and the films mask characteristics. This is pretty tricky to do, involving quite a bit of testing but the differences in the final result are like chalk and cheese and the resulting files are far easier to edit.

So I would probably be better off with filtering the flash head with some gel filters that can color correct for the mask? What kind of filters? I've glanced at some color enlarger heads on ebay and they seem to use a combination of yellow, magenta and cyan filters to balance out the light. Are those kinds of gels sufficient?

4) Despite the accepted wisdom that film only needs to be scanned at say 8 megapixels becuase it is only holding that much data the truth is you really need far higher levels of resolution to extract the information in the neg without alaising the grain. As an axample I have regularly scanned fuji 400 in sections for a final stitched output of about 100 megapixels, at this resolution you actually do resolve the grain of the film without alaising, the alasing causes the grain to look much larger than it really is and messes up many of the subtleties of the neg. Once fully edited the files are then downsampled to something more sensible like 16 megapixels and then print beutifully and look much more detailed than any other scanning method. Frankly even the best commercial scanners have insufficient resolution to really do the job properly. Lower speed films like Fuji Reala need around 200 mp or more at capture. I realize most people will think I am talking theough my hat but I promise as you go higher to these extremes the granularity of the film capture shrinks and becomes more tightly focused making for a subtly better final result. Ultimtely with scanning there are two approaches to resolution, either scan low enough so grain is smoothed out altogether ( which will greatly soften textural details etc but is fine for web and small prints) or scan high enough to sharply render all the grain in the film, anything in between is just messy.

I figured I would try this method only for the photos I love the most, thankfully my hobby as a film photographer was short. Right now I have been capturing slides and negatives at a 1:1.5 ratio(this is based on the marking of the full-frame macro lens I use) because of the size of my camera's sensor. Will going to 1:1 be sufficient for an APS-C sized sensor or should I go higher like 2:1?

5) When exposing go for a histogram that is roughly centered, exposing to the right does not help with neg scanning and underexposure will cause radical increases in image noise in the highlights of the inverted image, it is totally different to how we would work with normal digital capture. Small deviations from correct exposure can and will mess up the colour in the highlights or shadows and make editing much harder, you must use manual exposure methods for the capture and by far a filtered flash is the best light source but I have had great results with high powered and filtered LEDs.

That is what I was wondering. I have seen some websites say to expose to the right on the histogram, but when I look at the histogram in Silkypix, the red channel is clipped if that happens. It didn't sound right to me that losing some of that red information was good when inverting. So I have been exposing in the center because it was the only way I could safely assume all red, blue and green channels were not being clipped off.

There is way more I could say, but hopefully this gives those interested some food for thought and potential pathways to follow.

cheers

Brad Nichol

-- hide signature --

Trying to make the complex simple

Thanks it has been giving me ideas to try out. Once I get a rosco swatchbook in the mail I'll try out the flash filtering.

Nice to read that my thoughts are going in the right direction. See thread with a smaller digitization project here:
http://www.dpreview.com/forums/thread/3680724

It is about larger formats but it is still the film material (bw, color inv, color neg, doesn't matter).

My maximum will be macro 5:1 (about 20 kDPI) with the assumption of the resampling to the 4 kDPI. 25 times smaller resolution as the result versus initial digitizing.

Expectations are not only about details in drawing but higher bit depth result (in the best case in some larger color space). Especially so big hight bit dept that resampled result will be better than RAW format from any of the DSLRs.

ZOIP wrote:

On the issue of filtering it is difficult to say what would work best, I use proper CC Filters but they are hard to get and expensive, I suggest you start with a moderate cyan filter, hopefully the ones you get will include something in that ball park.

As for resolution, you don't need high res unless you are printing 8x10 or bigger and certainly for web use low res is probably preferable, but countless experiments and jobs I have done show high res stitched works brilliantly for high quality 11 x14 prints and bigger, the resolution bump give an increase in textural information and subtle tonality. It does not increase grain size, but rather resolves it better which actually makes it look smaller and tighter in "print", due to it not being alaised by the fixed pixel matrix of the cameras sensor. Note I refer to print, neg film is designed and intended to be printed, not for digital display, I think this is a sticking point when people make comparisons between film and digital, generally film will look grainy and rough on screen (there can be a messy interaction between the grain of the image and the fixed pixel grid of the monitor) but well scanned and processed colour neg images can print beautifully displaying tonal subtlety and texture that digital often misses, its all about the final print. The grain is actually an intrinsic and essential part of obtaining the look so removing or messing it up in some way will work against the look, but if all you need is on screen images then low res, (below the level where the grain even begins to be rendered at all) is the way to go.

Ulimately it comes down to artistic intent and purpose as to how you approach the resolution question.

On the exposure bit, yes middle is best, clipping the red channel or any channel really messes things up when you invert the image making proper colour and tonal correction near impossible, the idea of pre-filtering is to keep the exposure across the three channels in proper balance . Red channel clipping can mess up your skies for example because the red channel when inverted is cyan, it pays to consider that we are dealing with a negative process.

Hope it helps

I took one of my copies to a print shop that did 8x10, and for the most part I was satisfied with the results of a 16mp copy.

I had to check how much a cyan CC filter costs, not cheap. Though there seems to be some good alternatives on ebay. I'm expecting a used cyan filter to arrive in the mail this week. For now I have a blue 80A CC filter, and a "variblue" filter which more or less puts out similar light, though I would say that variblue filter is closer to the cyan scale.

Putting the filter on the camera lens did reduce the sharpness by a tad, though some digital sharpening filters seemed to offset it.

I have a rubber lens hoods and just now thought to my self, if I just put this over the flash head with a filter attached to it, it's basically like a flash gel. Here is the result of that using the 80A CC filter.

Only did an invert and some white balancing in software, but no other touch ups.