UV Induced Infrared Fluorescence 2

[Stefano]

I don't know if I understand. The images are UVIIRF, so you can't white balance on PTFE or similar, and to WB you use cotton. And you can't see red with the 720 filter (Hoya R72?) but you can with other filters? If so it is possible, but I don't know If I understood what you meant.

 

If you take a normal reflected IR image with a 720 nm filter, and you white balance it off a neutral target (PTFE, paper, etc.), the rainbow of false colors you get starts from an orangish yellow (shorter wavelengths, in the far red region), then white (there is always a neutral point having essentially two channels), and then a cyan blue for the longest wavelengths, like 900 nm. If you white balance your images off something else (like the IR fluorescence of cotton), you may get other colors.

 

Then, as David said, you don't have to be always scientifically accurate, if you like how your images come out, then they are Ok.

[Ninjin]

Well, see example. This is an 700+, IR photo (right), but used 1 faint LED near the top of the flower. This makes "red." (although the leaves are blue) This gives a fake red on top of the flower. Can be used like that.

 

 

Or here's an example. This flower is colored red-orange. The same effect. This is an 715 (I forgot exactly) IR photo but I used a fake red for decoration just by forcing it.

 

 

So fake red may well be in the IR, I think. This is all accurate. If you put the light on the same way, you will get the same thing. Even without any flags.

[Stefano]

Three things:

- Are these UVIIRF images?

- How did you white balance them?

- What kind of LED did you use (wavelength)?

[Ninjin]

Three things:

- Are these UVIIRF images?

- How did you white balance them?

- What kind of LED did you use (wavelength)?

 

- No, this is an IR photo (reflected IR). As example, I am showing that this is possible in an IR photo.

IF an object emits IRF in THIS range (as for example, a Ruby stone), it will be the same color, red.

- Balance - The gray zone in NX 2 (is always the same).

- Led just one from Ikea (a household LED on a flexible white leg) that screams about 700-730 nm like abnormal.

 

 

If you want, take the any plant, put two 60w incandescent lamps, near.

And one LED about 700-730 nm + _ Directly in the flower. (Yes, any household LED can work).

 

You will get the same. Red flower, white - blue leaves. In one photo. The camera needs a filter 715 or slightly lower than 715. Just a little. I like use the old K19 for, but probably there are better analogs, I'm not sure.

 

See, I tried to draw about where it is all, but I'm not sure if I wrote it correctly.

[Stefano]

Maybe I started to understand. If you white balance IR with a 720 nm filter in daylight, an incandescent light bulb will appear bluish, since it peaks at longer wavelengths than sunlight. This means that the light you used is bluer than "normal". Since you white balanced under this bluer light, false yellow objects become redder.

 

Try this: when you have the sun (it should be nighttime for you now), around midday, try to see sunlight with your camera with your 720 nm filter with an in-camera white balance set on those incandescent light bulbs. Sunlight should appear yellowish, and everything should have a yellow color cast.

[Cadmium]

Ninjin, Very nice!

By '720' filter, do you mean R72 or RG9 filter?

 

Stefano, the golden fruit is UVIIRF ('LUM').

[colinbm]

Colin - unfortunately I don't know which colour temperature it offers. This is what I bought, but I can't see that info. in the description:

 

https://www.amazon.c...0?ie=UTF8&psc=1

 

It claims to use the Cree "Improved XM-L L2 LED chip (no other cheap small XM-L T6 LED chip)". I'm assuming this means the XM-L2. You can get the data sheet from here:

 

https://www.cree.com...ete/xlamp-xm-l2

 

There are a number of different versions of that LED for different colour temperature ranges, but I don't know which one is in the Lumitact. I guess you could ask a local reseller or the AsiaPac distributor (my "manual" only gives the EU distributor).

 

Thanks Bernard

[Stefano]

Ninjin, Very nice!

By '720' filter, do you mean R72 or RG9 filter?

 

Stefano, the golden fruit is UVIIRF ('LUM').

Yes, I was referred to the last images.

 

If the filter is RG9, does it make a noticeable difference? How do R72 and RG9 compare in actual images?

[Ninjin]

Stefano, I think not like that. Range is important, the balance is not so important. Range = constant color. It doesn't matter what the light source is.

The emitter/re-emitter can be the Sun, Led, flower, or even a 'dry panda?'.

Only the range is important. 690-700 will be red from any source, won't it.

 

See example UVIIRF (IR luminescence):

 

The source of light (re-emitter) 600-700 nm - are the stamens of the flower.

IF I want to photograph color of this range I have to take a filter of 700 nm or lower. Then I see the light Red-false orange from the stamens. Photo on the left.

But If I took the 715 filter - photo on the right, I cannot see color of this radiation. Because I took a step of 15 nm. To see it again - I have to step back 15 nm.

So this color is exact thing, even 15 nm matters.

 

 

Tell me if you see any mistake here.

 

UVIIRF (IR luminescence) 610+

 

[Stefano]

A sensor sees always the same color at a certain wavelength (if you don't overexpose), but when you white balance you alter those colors, sometimes quite a bit.

 

Anyway, I will try this too and post the results.

[ulf]

Ulf, not sure I've ever mentioned about a 2E, have I? I like Zeiss T* for blocking (or my KV418, or even a 420nm longpass I got from Klaus) as they all have good blocking with about 50% max transmission in that 415-420nm region, and low fluorescence. I've not tried 2E.

Sorry Jonathan, I included you as KV418 and the other long-pass filters you mention also cut the last of the visual range just as the 2E-filter.

 

I am just trying to understand why that part of the visual spectrum is considered harmful for fluorescence photography.

Without any personal experience with this I am trying to learn.

 

From what I can see from my measurements there is very little risk of UV-contamination from the light-source with a Nemo with it's original filter and my Tiffen Haze 2A-filter on the lens.

[JMC]

Ulf, for me it comes back to being certain about the areas of the spectrum being blocked. Filters start blocking a good few nm before their 50% value, so having something with a 50% max transmission around 420nm allows you to be pretty certain that it is blocking everything above 400nm. KV-418 is(was) commonly used as a blocking filter for UV fluorescence work due to its low fluorescence and its cut off wavelength.

[ulf]

OK. I understand the principle.

Block the light source and camera lens in a way that eliminates any overlap.

 

This discussion is about having very big margins.

It is like putting your piggy bank in a safe and keeping the safe in a bank vault.

For me that is overkill, but i might be missing understanding something essential.

 

Or the images taken with those filters are also missing some of the blue and violet fluorescence that I am convinced also exist from some types of materials.

 

Why is the fluorescence of the last 20nm above 400nm so very uninteresting?

[Stefano]

A sensor sees always the same color at a certain wavelength (if you don't overexpose), but when you white balance you alter those colors, sometimes quite a bit.

 

Anyway, I will try this too and post the results.

Replying to a previous conversation (should we split it off?), this is what I got. I used my Panasonic DMC-F3 with a Hoya R72 for all images. I used the 5 W filament of a car stop light, running it at a crazy 32 V (it takes I think 39 V to break the filament), and took a picture of its spectrum with a 1000 lines/mm diffraction grating, with an in-camera white balance done on a paper sheet.

 

Result:

 

As you can see, I have some orange, some greener yellow and a lot of blue, plus the white point.

 

This is how the spot of a white LED torch looked line on a wall, with the same white balance:

 

To me, that's a shade of yellow, a bit on the orange side, but still yellow.

 

Then, I repeated the above test but this time I white balanced on the "blue" 940 nm light coming from a remote.

 

Diffraction grating:

 

White LED:

 

This time there was more orange, but I still didn't get your reddish oranges. Maybe your camera sensor just sees light in a different way.

[Andy Perrin]

Stefano, your results and Ninjins are consistent. She used 700nm long pass and 715nm long pass. You used 720nm long pass. She demonstrated that 715nm cuts out the red so of course 720 will also.

 

I would say that you and Ninjin are arguing about the same issue as Ulf and Jonathan — where do we put the cutoff point for fluorescence? On either end of the spectrum!

[Ninjin]

Stefano, thank you very much, I am very interested in your result!

I think everything is correct in your result, you showed everything as it is.

The second option seems to me almost the same as mine.

 

Yes, I have a 'bit' more reddish tone on the my diffraction grating. I've tried 2 different cameras and both show the same for all my filters.

Why there is some small difference with your test - I cannot say right now. There may be a different filter. Or balance after all. Need to see more.

Thanks again for this test! I must try to do something else.

I have a comparison of 715 and RG9, but I will make a topic about this separately then. Perhaps, it would be more appropriate.

[Ninjin]

Cadmium, thank you! I think it was RG9. 715 works same.

[dabateman]

For limits always consultthe arbitrary, but specific government regulations.

Like this for UV:

https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfCFR/CFRSearch.cfm?fr=201.327

 

Solar UV is from 290nm to 400nm

UVA II is 320nm to 340nm

UVA I is 340nm to 400nm.

 

 

From here infrared is 700nm to 50000 nm

 

https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfcfr/cfrsearch.cfm?fr=890.5500

 

[Stefano]

So visible light is defined here as I do, 400-700 nm. IR is quite a broad band, 50 μm is well into thermal IR and about the diameter of a human hair. The frequency is 6 THz, so it borders with Terahertz waves.

[Ninjin]

If I understand correctly, you are saying that red is not considered a false color.

 

If I said - "I can get false red as 1 color of infrared luminescence". For you this is not correct since red - is visible (not false)? This is if I understood correctly.

 

Ok, maybe you're right.

I said so - because I count visually. And visually - Red for me ends where S8612 1 mm cuts it. Or 2 mm BG40 which is the same.

For me, this is the objective end/border of visible red for any photography.

 

This end/border can be clearly seen on the diffraction grating. Anything further - it is invisible red. Because the S8612 1 mm of the visible picture limits it.

 

Then, if I take a photo of the visible range of S8612_1 mm, on a diffraction grating (end of the red tail).

And then, for example, a photo of the range on the diffraction grating of your 720 filter.

 

Between the end of the visible red tail S861_1 mm and the start your IR filter 720. Between them there will be a section of objectively red color.

What should I consider in this red color part? Yes, in my 'personal light system', this is false red, since the visible range has already been completed/end.

 

Ок, probably I should not say 'false red' but 'invisible red'. Perhaps this is more correct, but - I'm not sure.

[Ninjin]

This is what I mean. Several different filters options on this one 1000 diffraction grating.

 

My question is about the part with green arrows.

 

This is already cut off by a S8612. Also, the any camera with hot mirror no longer sees it.

 

But it is not yet 'false orange infrared'.

So how should I name this range correctly?

 

All the most beautiful flower stamens are here.

 

[Cadmium]

Ninjin, Nice comparison, fascinating!

Do you have one of those showing an RG715?

[colinbm]

Ninjin, unfortunately our cameras are stuck in the 'Colour Wheel', MRYGCBM........(RGB + CMY), Magenta not being a colour of the rainbow, but the link in the colour wheel to make red & blue mix & meet.

Our cameras can record wavelengths from near 300nm to near 1000nm, but not the true colours except RYGCB, that is 680nm to 420nm.

Our cameras can't see the Violet & Indigo nor the richer reds before IR.

I would be nice if our cameras could see the full range of human visible colours.

[Stefano]

With "false red" I meant a red color visible in a white balanced photo that isn't actually red in real life. An artificial red. This could correspond to IR or other colors. For example, deep IR appears blue in a typical white balanced image, but that doesn't mean deep IR is actually blue, it is every color and no color at the same time, you are free to choose.

 

I have a far red LED, peaking at 730-740 nm. I can see it quite well, although it is much dimmer than my 660 nm deep red LED.

 

The color this LED has... it's insane. It is as red as it gets. The reddest red you will ever see. No screen can reproduce it. It is even slightly redder than the already very red 660 nm one.

[Andy Perrin]

Ninjin, unfortunately our cameras are stuck in the 'Colour Wheel', MRYGCBM........(RGB + CMY), Magenta not being a colour of the rainbow, but the link in the colour wheel to make red & blue mix & meet. Our cameras can record wavelengths from near 300nm to near 1000nm, but not the true colours except RYGCB, that is 680nm to 420nm. Our cameras can't see the Violet & Indigo nor the richer reds before IR. I would be nice if our cameras could see the full range of human visible colours.

Colin, the problem isn't the camera, the camera (as you said here) responds to indigo and violet wavelengths just fine. It's that the MONITOR can't display 400nm using an LED that peaks at 450nm.