School me on color checkers for scanning

bernard_L said:

Can you elaborate? Did you find in the scene objects that you thought/knew were neutral grey, neutral white?

Or did you shoot one frame with a proper white/grey/black card and used that as a reference --90% of the effectiveness of a ColorChecker for 10% of the cost. ??

Click to expand...

Thanks Bernard, it was the former. And I’m glad you mentioned the white/grey/black cards because I forgot to mention that I ordered those when I returned the color checker. That’ll be the next experiment.

warden said:

Thanks Bernard, it was the former. And I’m glad you mentioned the white/grey/black cards because I forgot to mention that I ordered those when I returned the color checker. That’ll be the next experiment.

Click to expand...

Definitely better than guesstimating what is a neutral grey. The only neutral color I'd trust in a scene is a fluffy fair-weather (not stormy weather) cumulus as a white point. Something else: do not use white paper as white point; because of optical brighteners, it will ruin your calibration. As black point I'd take film base.
With the grey card, beware of specular reflections, i.e. mentally replace the card with a mirror: is the sun's reflection in the field of view of the camera? And stay away from large colored areas, such as colored building, foliage, etc.

<joke> The iphone pictures are visibly better. We want vivid colors. Ditch your camera. </joke>

koraks said:

that clearly didn't pan out very well.

Click to expand...

What do you mean? You don't like my color? I'm insulted!

koraks said:

Just to be sure, this is the setting I use for this purpose on my old 4990:

Click to expand...

Yeah me too, looks like this:

And the resulting scans look like:


btw I forgot to include the image below earlier but if I just let Epson scan do all the thinking in auto mode the results are much closer to what I see on the iPhone image, and also closer to my memory of the scene:

I'll continue the experiment when the white/grey/black cards when they arrive. In the meantime if anyone wants to play with one of my base scans I'll be happy to send one.

bernard_L said:

<joke> The iphone pictures are visibly better. We want vivid colors. Ditch your camera. </joke>

Click to expand...

Or go back to TMax and be happy!

Thanks for all the advice Bernard.

warden said:

btw I forgot to include the image below earlier but if I just let Epson scan do all the thinking in auto mode

Click to expand...

Yes, that works well especially on an image that overall balances out to pretty much middle gray. It doesn't work well if there's a strong/dominant color in a large part of the frame; that's what went wrong in the Prayer House shot. And this illustrates that the scanner/software will come up with a different balance for each shot, which can be rather annoying when working on a series photographed under similar conditions.

warden said:

What do you mean? You don't like my color?

Click to expand...

Haha, oh no, your color is absolutely fine!
In fact, it's really not so horrible at all save for that one frame, and the slight inconsistencies of the series as a whole.
Btw, this is what I get if I take your negative and run it through my 'process' (i.e. manual inversion + adjustment of the R, G and B curves):

From here, I would usually fine tune it a bit to taste with a final adjustment curve. This curve I would then also apply to the other images of the same series to maintain consistency.
For instance, in this case, I would opt to make the midtones a little less green, the shadows a little less cyan and add a little blue to the highlights to approximate how I would subjectively experience a scene like this:

koraks said:

Yes, that works well especially on an image that overall balances out to pretty much middle gray. It doesn't work well if there's a strong/dominant color in a large part of the frame; that's what went wrong in the Prayer House shot. And this illustrates that the scanner/software will come up with a different balance for each shot, which can be rather annoying when working on a series photographed under similar conditions.

Click to expand...

Exactly. I want to color correct for all shots in a given lighting situation rather than one image at a time so there is consistency. I’d rather start with a base scan, post process it and save/reuse those curves for all similar lighting conditions on that day’s roll of film.

And I couldn’t choose a neutral grey in a scene if my life depended on it so I’m looking forward to the next experiment using cards. Thanks for your help!

Here's an update. I think I'm making progress and I'm enjoying the experiments.

This result is better than I was getting before and is approaching a quality that I can use. I'm still experimenting with my approach, which is relying on a combination of the white/grey cards, using both Lab and RGB color space, and having a cell phone pic of the scene handy to check spot colors for obvious problems that my eyes might not catch. In this case I checked the four spots listed just to see how close I was to what the phone saw (which I know is a processed image but still). (Choosing what to sample is a challenge.) All four are fairly close but I'm sure you'll spot the differences. Without the combo of the white card and cell pic I'd be lost for sure.

I'm learning that the phone over-saturates skin tones when I use the Lab guidelines that @Mr Bill discussed earlier in this thread so de-saturating the phone pics a bit before relying on them for a disaster check tool is needed. Lab color is really helpful if like me you don't see color well.

I did an experiment this week where I took several pics of my bag in differing lighting conditions and tried to make it read correctly in full sun, full shade, etc. That was a good exercise. Thanks to @Mr Bill, @bernard_L and everyone else for the assistance. I appreciate it!

Overall, image saturation is a challenge that will need more experimentation. I'm getting images that are on a spectrum between dull and candy colored and it's hard to get it right. I'll keep at it and will appreciate any feedback you might have.







Here's what the phone saw:

Looks good; you've got some blue/yellow crossover in the film shot (well, a lot actually), but I personally don't find it bothersome. You can spot measure the hue on the black leather case just below the number "4"; notice how it's close to neutral (probably very slightly yellow/tan) on the phone shot, but very blue on the film shot. The same goes for the black straps on the backpack etc. The film shot is balanced more towards blue on average on the whole, but it's more pronounced in the shadows.

There's also a little magenta/green crossover with green shadows in the film shot. It's visible in the vertical concrete face where you placed the number "4". Again, ask yourself if it's a problem, per se.

koraks said:

Looks good; you've got some blue/yellow crossover in the film shot (well, a lot actually), but I personally don't find it bothersome. You can spot measure the hue on the black leather case just below the number "4"; notice how it's close to neutral (probably very slightly yellow/tan) on the phone shot, but very blue on the film shot. The same goes for the black straps on the backpack etc. The film shot is balanced more towards blue on average on the whole, but it's more pronounced in the shadows.

There's also a little magenta/green crossover with green shadows in the film shot. It's visible in the vertical concrete face where you placed the number "4". Again, ask yourself if it's a problem, per se.

Click to expand...

Thanks for the feedback Koraks. One step at a time. For now I’m asking “what part of the image is most important?”. The answer is “all of it” lol.

warden said:

For now I’m asking “what part of the image is most important?”. The answer is “all of it” lol.

Click to expand...

Hehe, yeah, I think that's the basic challenge with photography!

Ok, one more update because I'm still practicing and you're still reading this.

I'm on the lookout for color crossover in the shadows and I think I'm spotting it better now. I'm aiming for a repeatable process and I'm getting close, at least I think so. My image editing time is going down and the quality is going up. But please let me know if anything looks weird to you. One side effect of this process is that I'm dreading color a little less. So that's something.

btw, the other images in this thread were made with expired but always frozen Portra 160. I wanted to use some of my old stuff while I worked on a scan/process technique. These two are made with Portra that I purchased a few weeks ago. Apparently B&H are selling short dated Portra for some reason as this stuff will hit its "best used by" date in 2024.

warden said:

I'm on the lookout for color crossover in the shadows

Click to expand...

You mean just from looking at the image on-screen and judging whether it's right or not? I would not trust such a method (and my eyes) myself. You stated the same previously:

warden said:

And I couldn’t choose a neutral grey in a scene if my life depended on it so I’m looking forward to the next experiment using cards.

Click to expand...

Plus, this lumps two causes of "visual crossover": film curves, and the shadows being (under sunny conditions) lit by blue sky. Whether to correct for that is debatable, but has nothing to do with film crossover.

bernard_L said:

You mean just from looking at the image on-screen and judging whether it's right or not? I would not trust such a method (and my eyes) myself. You stated the same previously:

Plus, this lumps two causes of "visual crossover": film curves, and the shadows being (under sunny conditions) lit by blue sky. Whether to correct for that is debatable, but has nothing to do with film crossover.

Click to expand...

I think I used the wrong terminology there Bernard. I just mean overly blue shadows, which I hadn't caught before but that I am noticing now, both in the image and also with curve adjustments that seem too extreme, if that makes sense.