• Ultraviolet Photography
  •  

Found a colorchecker target designed for UV Fluorescence Photography

Fluorescence White Balance Processing
13 replies to this topic

#1 ultrainfra

    Member

  • Members
  • 22 posts

Posted 26 January 2021 - 01:35

I searched through the forums for this, as well as searching for the URL at which the product is located, so seems like no one has shared this before. I figured I'd bring it to everyone's attention.

It only has RGB and GBW squares, and costs a whopping $875 (US dollars) though. :omy:

http://www.imagescie...ry_Code=TARGETS

#2 Andy Perrin

    Member

  • Members+G
  • 4,084 posts
  • Location: United States

Posted 26 January 2021 - 02:46

Not sure I’ve seen that particular one but I knew they existed. The problem has always been that they are super expensive.

#3 dabateman

    Da Bateman

  • Members+G
  • 2,767 posts
  • Location: Maryland

Posted 26 January 2021 - 06:30

Well since its optimized for 2E to Peca 918, its actually $1225. As you will need to use those filters on the camera that they have developed the target for, which they sell.
Thats also part of the problem with standards for UVIVF, as if you use a BG38 or S8612 or whatever as your favorite IR cut the reds will be significantly different. Sometimes I like more or less red depending on the subject. So I don't really like a standard and thus dropped the idea of using one.

Its only a standard if everyone in the world has agreed to those cut offs for Blue and Red. I don't know if I fully agree.

This is their FAQ describing why their standard isn't working for you:
https://www.uvinnovations.com/faq

This is what their standard has been developed for:
https://static.wixst...02573b~mv2.webp

I also forgot about excitation. Apparently it has been developed for 368nm UVA light only. So if you use a 385nm or 372nm light source you will not be in the standard range.
As you can see from Andy's recent pumpkin images and other induced fluorescent images, the excitation wavelengths will also affect the amount of reds and blues you get. 405nm, although not technically UV does give nice color.
Again why I have dropped the idea of a standard.

We need a group of people to sit in a room for a week, to come up with an ISO regulation. Then for sure I will not follow that standard.

Edited by dabateman, 26 January 2021 - 06:43.


#4 dabateman

    Da Bateman

  • Members+G
  • 2,767 posts
  • Location: Maryland

Posted 26 January 2021 - 06:55

Looks like IUPAC has some recommendations:
https://www.research...echnical_Report

Also looks like ISO has one now for confocal microscopes:
https://www.iso.org/...so:std:69820:en

Interesting ISO 4892-3 recommends 340nm as UVA source for weathering. The fluorescent lights that are discussed in ISO 4892-3 are: UVA-340, UVA-351, UVB-313 or a combination of four fluorescent lamps.

I found this description:
https://www.micomlab...-Test-Table.png

Edited by dabateman, 26 January 2021 - 07:06.


#5 Andrea B.

    Desert Dancer

  • Owner-Administrator
  • 8,762 posts
  • Location: UVP Western Division, Santa Fe, New Mexico

Posted 26 January 2021 - 07:16

There is a discussion in the UVIVF Sticky about the UV Innovations Target-UV™ and the UV-Grey™ which are sold through Image Science Associates. There are 4 links in the Sticky the review and tests I made for these targets. Scroll down to the section White Balance in UVIVF Photography to find the links.
They aren't perfect tests. I stumbled around a bit, but eventually got the hang of it.
<> Sticky :: UV Induced Visible Fluorescence <>

The UV-Grey is the most useful item because it allows you to set an in-camera white balance under the UV illumination. I think you can do without the Target-UV because we don't generally care about measuring the brightness (or amount?) of fluorescence.



The primary problem of UVIVF white balance is to find a working definition of exactly what "white" is in the context of UV-induced luminescence. For the most part, luminescence is narrowly excited and narrowly emitted. So how in the world do you make a "white" fluorescing target? Would the emitted light be a combo of all color wavelengths? Or would the emitted light be only R, G, B wavelengths? Yet the UV inducement of this "broad" emitted "white" light is constrained to a rather narrowband UV-LED peaking around 365 nm.
Kinda sounds impossible. :smile: So I'm fairly impressed by the UV Innovations effort.



I should remind everyone that our usual fluorescence photography here on UVP is quite outside the typical laboratory fluorescence work where narrowband excitation and emission filters are used and where the emitted light is usually recorded via spectrometric equipment. We are using the narrowband excitation but we are attempting a very broadband capture of the visible emissions. A side problem with this is that we are also capturing reflections of the emitted light off of parts of the subject which may not be fluorescent. (Hope I phrased that correctly.) I hope someday to try some narrowband captures myself and see some other folks try that too.



Added: The UV Innovations FAQ page is excellent. Worth reading even if you don't use their targets. Linked in David's post #3 above.
Andrea G. Blum
Often found hanging out with flowers & bees.

#6 Andy Perrin

    Member

  • Members+G
  • 4,084 posts
  • Location: United States

Posted 26 January 2021 - 07:40

Andrea, I noticed this went beyond just white balance with a full color checker (albeit non-standard layout, so it wouldn't work with PhotoNinja).

#7 dabateman

    Da Bateman

  • Members+G
  • 2,767 posts
  • Location: Maryland

Posted 26 January 2021 - 08:06

Frankly I am disappointed. If 340nm is the standard for UVA weathering, why aren't there more 340nm lights available. I would like a 340nm, 365nm and 385nm specific UVA lights to cover the range.
Colin was getting there with his 4 LED lights he was building. But I think I want them isolated to look for difference.

View PostAndrea B., on 26 January 2021 - 07:16, said:

The primary problem of UVIVF white balance is to find a working definition of exactly what "white" is in the context of UV-induced luminescence. For the most part, luminescence is narrowly excited and narrowly emitted. So how in the world do you make a "white" fluorescing target? Would the emitted light be a combo of all color wavelengths? Or would the emitted light be only R, G, B wavelengths? Yet the UV inducement of this "broad" emitted "white" light is constrained to a rather narrowband UV-LED peaking around 365 nm.
Kinda sounds impossible. :smile: So I'm fairly impressed by the UV Innovations effort.


Thats what a group of people in a room for a week decide. It doesn't have to be based on anything and usually always has some issues. Just like most standards.

Edited by dabateman, 26 January 2021 - 08:11.


#8 Andy Perrin

    Member

  • Members+G
  • 4,084 posts
  • Location: United States

Posted 26 January 2021 - 08:42

So, the fluorescence spectrum can be arbitrarily wide relative to the inducing light source’s spectrum— extreme example would be the laser-induced fluorescence I’ve recently posted to the board. That’s very very sharply peaked at 405nm or so (lasers shift a bit just like LEDs) and the emission spectrum is all the way out into infrared (in fact testing for laser-induced SWIR fluorescence is on my list of things to get around to).

#9 dabateman

    Da Bateman

  • Members+G
  • 2,767 posts
  • Location: Maryland

Posted 26 January 2021 - 09:07

Yes a standard for UVIVF would have to specify the specific light source used for excitation, the specific visible cut detection area (including intensities at each wavelength) and the detector sensor. Ideally a monochrome sensor would be used and specific specified blue, green, and red filters with tight cuts would be included in the specification. However I don't think a manufacturer sitting on the board in that room would fully agree and a subsection would need to be added for GRBG sensor with the standard Fuji color filtration dyes. A Canon member might argue for inclusion of their dye system with differences in green channels, if OM systems is around (Olympus) they might agree as their two greens are also slightly different.

I have been there, done that in different world.
The duration of illumination would also most likely need to be specified with the storage conditions for the sample.

Now we can see why its more fun to just photograph stuff and tune the colors to our liking.

Hey look this is kinda green. Verse this sample of X is always candy apple green. I know I have X present as I see the candy apple green color.

Edited by dabateman, 26 January 2021 - 09:10.


#10 Andrea B.

    Desert Dancer

  • Owner-Administrator
  • 8,762 posts
  • Location: UVP Western Division, Santa Fe, New Mexico

Posted 26 January 2021 - 19:31

I think, though. that the standards you are describing are probably overkill for our simple photography. But your background is very science laboratory oriented, of course.

Back when I tested the Target-UV and Gray-UV targets my main goal was to try to find which in-camera white balance might "match" the results those targets gave. The closest I came with a Nikon was to set K10000. I should really retest with the S1R to determine whether it is possible to get closer.

Also I think my original tests were not so great because that flower I used was not very fluorescent-y.
Andrea G. Blum
Often found hanging out with flowers & bees.

#11 JMC

    Member

  • Members+G
  • 1,270 posts
  • Location: London, UK

Posted 29 January 2021 - 14:56

If it's of interest, I found a couple of historical documents regarding UV fluorescence calibration standards.

First one, it's one page (p155, called "Sintered mixtures of phosphors in polytetrafluoroethylene resin for fluorescence standards") which talks about testing of some NBS colour standards during fluorescence - https://www.govinfo....8b490aa0cd0.pdf

The paper above also goes into the details on how they are made. It also references a second article, and the link to that is here - https://www.govinfo....120dc83a097.pdf
Jonathan M. Crowther

http://jmcscientificconsulting.com

#12 Andrea B.

    Desert Dancer

  • Owner-Administrator
  • 8,762 posts
  • Location: UVP Western Division, Santa Fe, New Mexico

Posted 29 January 2021 - 16:48

Thanks, Jonathan.

I'm hoping someone here will experiment with a mix of fluorescent paints and perhaps get close to a fluorescent white.

Somewhere I mentioned that I told a LabSphere representative that a white fluorescent standard would probably be a good seller for the art preservation crowd and for a few UVIVF photographers.
Andrea G. Blum
Often found hanging out with flowers & bees.

#13 OlDoinyo

    Member

  • Members(+)
  • 864 posts
  • Location: North Carolina

Posted 04 February 2021 - 20:10

It was mentioned in another post here that some mineral specimens (at least under some excitation profiles) emit something like white light. If a standard phosphor is desired, would powdered rock be a suitable candidate?

#14 Stefano

    Member

  • Members(+)
  • 1,796 posts
  • Location: Italy

Posted 05 February 2021 - 08:47

Maybe, but which rock? Natural rocks have some variability. Mixing highlighters in a certain amount may "create" white, but it would still be hard to standardize the process. And given the fact that even defining white is not as easy at it seems, it is very hard to make a target like that.