Matching lighting of two pictures for a scientific research project

[mvnural@gmail.com]

Hi,

We do research on the effect of fish food on the color change of the fish's skin. Therefore we observe the fish over multiple weeks and take their picture in our studio environment once a week. In order to track the color change of the fish, all other conditions (lighting, distance etc.) should be exactly the same so that we can compare the average color of the fish over multiple weeks.

The problem is as follows. We have created an studio environment with a light tent and three flashes and fixed them to keep the shooting conditions the same every time. However as you can see in the three images below which are shots of the same fish in three different weeks, the lighting is not the same. We think that since we do shoot 30 fish in each batch, the flash did not produce the same performance for every shot and we got different lighting in each shot.

How can we match the lighting of these shots so that we can compare the color of the fish? We already tried the color match tool however it is no use in this case. Since we want to keep the fish as it is and only adjust the slight lighting difference.

Thanks in advance for your replies.

 

[RTContent]

The problem you have is white balance/exposure. The problem may not be the flash but the camera that you are using. If you are using an automatic setting on the camera, it will meter the scene on every shot and your exposures will differ slightly (giving you a slightly differently white balance).

This can be adjusted in Photoshop though. Just go to Image - Adjustments - Exposure. Click on the lightest eyedropper icon and then select the white sample from your palette.

Hope that answers your question.

 

[sprucemagoo1]

Why not replace the flash with a constant light source, like a spot light or two?

 

[NemesisWorks]

Why not replace the flash with a constant light source, like a spot light or two?

Agree

 

[ibis]

why Nemo?
Why???
Seriously, if not able to achieve same photographing conditions (that is almost impossible to have, but in strictly controlled environment) due to the multiple factors such as main object color change, battery life etc...
i did basic levels correction on bouth of the images add here are results, and i think fish color diffrence is visible. also those were low res images. hope something like this will fit you.

 

[mvnural@gmail.com]

Thank you very much. But I have tens of other fish shots waiting for me to be corrected. Could you please tell me step-by-step instructions for how you did the levels correction?

I would really appreciate that since I'm a PS novice. Also the shots are in low-res because of the size restrictions of the forum. I had to resize them that's why.

 

[ibis]

it is ok to resize images down in order to post them on forum.
if i have used full size images results will be better.
this works best on decent contrasted images.
open your image
hit ctrl+L

there is pipette with minus sign on it like on screenshot
just click now in avearge gray area
experiment with the source click point

 

[mvnural@gmail.com]

Thanks a lot. I think that will do it.

 

[ibis]

Thanks a lot. I think that will do it.

glad if helped.
just pay attention on the images quality in order to get it right. your last photo seems to be the quality contrasted kind.
ps if it went too far there is an option:
edit-fade levels

 

[ibclare]

Another corrective option after making the levels changes: usually it takes only a slight adjusting of the middle slider to get your images to match up, such as to get Nemo the same shade and color. Levels is a pretty quick method to achieve the results you want.

Often the white balance under exposure is a quick way, but may still require adjusting the middle grey range. You'll generally find those slight tweaks necessary no matter which method you use.

 

[mvnural@gmail.com]

Another corrective option after making the levels changes: usually it takes only a slight adjusting of the middle slider to get your images to match up, such as to get Nemo the same shade and color. Levels is a pretty quick method to achieve the results you want.

Often the white balance under exposure is a quick way, but may still require adjusting the middle grey range. You'll generally find those slight tweaks necessary no matter which method you use.

I don't know if I understood incorrectly but I don't want to get Nemo's of different shot the same shade and color. Because Nemo's are expected to change color every week because of their diet. And we want to measure that change. Therefore in order to measure that accurately, I try to match everything else except nemo.

I don't know if this process also tries to match the color of nemo in the source and target.

What do you think?

 

[ibclare]

OK, then white balance or levels if you don't tweak anything except to match the petrie dish or background. Either will work.

 

[ibis]

just do the white ballance, we are not trying to fix it to look good leveled.
fish color changes are visible after just a leveling, that will unveil the gray cover.

 

[Stric9]

I still think that something is off. If you look at this screen capture of what ibis posted, you can see that the respective swatches give diffrent results in them.
Now if the everything was constant and the white balance was corrected they should give the same color right?


I would think you would have to somehow bring all the color swatches into agreement with each other, to then be able to obectively test the colors of the fish. Just my thoughts.

 

[MikeMc]

Why not include an 18% grey card in each shot....makes it easy to get the same exposure.

 

[ibis]

I still think that something is off. If you look at this screen capture of what ibis posted, you can see that the respective swatches give diffrent results in them.
Now if the everything was constant and the white balance was corrected they should give the same color right?
View attachment 21763View attachment 21764

I would think you would have to somehow bring all the color swatches into agreement with each other, to then be able to obectively test the colors of the fish. Just my thoughts.

couldn't agree more.
what i did was just one click, in order to show how things work.
reference palette is there to do it's thing - being an reference. you need to experiment with source click point , to get the right shade of gray to get them even. reference palette-not the fish. then you'll be able to read fish color difference.

 

[Paul]

You could always gif it and just adjust hue for each frame

 

[MikeMc]

couldn't agree more.
what i did was just one click, in order to show how things work.
reference palette is there to do it's thing - being an reference. you need to experiment with source click point , to get the right shade of gray to get them even. reference palette-not the fish. then you'll be able to read fish color difference.

Scott Kelbys books have a 18% card for balancing RAW photos....if the 18% is the same on all photos, you are done, and the color variations will be the fish...Why are you trying to get the color balance equal?