¿ªÔÆÌåÓý

ctrl + shift + ? for shortcuts
© 2025 Groups.io
  1. QuadToneRIP
  2. Topics

Measure test prints with digital camera

 

I've been reading about using QTR to make a DN for darkroom?prints (silver gelatin). After making the test print (or step wedge), it appears that a flatbed scanner can be used to scan the print and measure the resulting digital file with Photoshop. We don't have access to a proper densitometer or other spectrometer devices.

Has anyone tried using a digital camera (DSLR or mirrorless) to?take a picture of the test print and measure the file in PS? As long as there is a maximum black patch and white patch, it seems these act as an internal standard for the image file. This could be used to compensate for exposure and lighting.

I'm sure someone has done this before - I just can't find any mention of this. Any advice would be appreciated!

Thanks,
RC Wieboldt
 

¿ªÔÆÌåÓý

There is a photoshop script that performs this operation. ?The script requires the use of a known gray scale reference as part of the scanned image to feed the script L reference values that are used to perform adjustment and?reading of the step wedge pattern you printed in your test print.

See:


And download the "QTR Stepwedge Tool: QTR-StepWedge-Tool.zip¡±

Then follow the documentation, included with the download of the .jsx file that gets installed in the Scripts folder of PS.

I have used the gray step wedge contained in the Tiffen Q-13 Small Color Separation Guide:


The gist is, you scan or shoot your test print and the gray Tiffen step wedge together and bring the resulting image into PS. ?The script will prompt you to select reference patches on the Tiffen target, modify your scan to bring it into conformity with these patches (via a Levels adjustment), and then prompt you to select your test print step wedge. ?It will then sample the patches in your test print and output L values for the 21 patches to a text file.

I suppose you could kludge something manually with a known target (perhaps a CC target) and use that target to set black, white, and gamma and then sample your test step wedge print manually, reading off the L values (set up the Info panel for 32 bit L for greater precision).

I suppose that you really want the scanned or shot image to be devoid of any Automagic adjustments so that you can measure the uniform gray step wedges without a contrast curve or some other monkey business injected into your process, so factor that in. ?I¡¯ve only done the scanner route, with an Epson V850 running Silverfast with all auto adjustments disabled. ?But I suppose the linearizing step of the QTR process would handle any funkiness in your step wedge data. ?

Take what I am saying with a grain of salt, I am not super experienced with all of this outside of my specific processes (positive inkjet prints and DN for silver gelatin printing based ?on Pictorico film DNs). ?I hope others more knowledgable than me chime in. ?

Good luck!

Kirk


On Apr 21, 2025, at 10:17?AM, PhotoMidwest Print Lab via groups.io <printlab@...> wrote:

I've been reading about using QTR to make a DN for darkroom?prints (silver gelatin). After making the test print (or step wedge), it appears that a flatbed scanner can be used to scan the print and measure the resulting digital file with Photoshop. We don't have access to a proper densitometer or other spectrometer devices.

Has anyone tried using a digital camera (DSLR or mirrorless) to?take a picture of the test print and measure the file in PS? As long as there is a maximum black patch and white patch, it seems these act as an internal standard for the image file. This could be used to compensate for exposure and lighting.

I'm sure someone has done this before - I just can't find any mention of this. Any advice would be appreciated!

Thanks,
RC Wieboldt
 
Edited

Yes, others (including myself, many years ago) have used a camera, but I do not recommend it for reasons given below. To be clear, any print-based linearization method, not just the QTR methods, face the same choices as to how to measure prints.
?
First of all, you didn't say if you have a scanner. Even an inexpensive scanner will give you better results than a camera. Do you have a scanner?
?
A scanner guarantees virtually uniform illumination. A camera + lights, unless you go to extreme lengths, does not. You say "This could be used to compensate for exposure and lighting." You can't compensate for uneven illumination. If you insist on a camera, I suggest getting one of the folding photo "cubes" used by people who photograph small items to sell on eBay. Get one with lights built in. I have one that I use for photographing finished prints (I don't have a scanner), so I can tell you that they're clumsy and take up space. A cube with all the bells and whistles will end up costing about the same as an inexpensive scanner.
?
If you are exposing your prints in an lightbox that also does not provide uniform illumination, then using a camera to measure your prints, unless you take serious precautions, simply compounds the problem. The symptom will be iterations that do not converge quickly to linearity. The problem will be even worse if you are making digital negatives for a high-contrast process.?
?
?
?
?
 

Thanks, Kirk! I somehow missed seeing this tool when we ordered QTR years ago. I tried it out and it is really slick. Also found this YouTube which does a nice job showing its use. It'll be a good resource for our members.
Appreciate the tip about the Tiffen wedge also.
Take care,
RC
 

Thanks for the reply, John. Your points are well taken.
?
Yes, we do have a couple scanners running VueScan and can go that route. It's just one more piece of equipment to train people how to use.?
?
RE my comment about compensating for exposure and lighting, I meant color temperature for lighting and camera exposure. Illumination is uniform so that isn't the issue. I'll try some experiments using window lighting with exposure off an 18% grey card and auto WB. When it's all said and done, you may be right that its just easier to use the scanner.
?
Take care,
RC
 

It isn't necessary for L values read with an eyedropper to be "true" L values, i.e., values you would get from a spectro. For linearization purposes, it is only necessary that the eyedropper L values have the same relationship to one another as the true L values.? You said it yourself in the OP: "As long as there is a maximum black patch and white patch, it seems these act as an internal standard for the image file."

1 - 6 of 6