6 replies to this topic

[Bernard Foot]

I've been dabbling with UVIVF and flowers, and I've been a bit disappointed. Apart from a bit of red chlorophyll and white dust, the results are not so different from a visible image. I guess it's ll about finding the right subject.

In the meantime I thought I'd see what UVIIRF threw up, together with the full-colour technique described elsewhere on UVP. Here are the initial results.

Excitation was a Nemo torch with a Baader U filter.

Camera for visible and UVIVF was an unmodified Canon EOS 6D II with a Tiffen 2A. For UVIIRF and IR, camera was a modified Sony A6000.

The tri-colour filters used for the three exposures were:

• MidOpt BP735 + R72 (CWL about 750nm) - becomes Blue channel
  
• MidOpt BN850 (CWL about 850nm) - becomes Green channel
  
• MidOpt LP1000 (CWL about 1000 nm) - becomes Red channel

A big question is how to do the WB. One thing that was clear was that the amount of fluorescence dropped off with wavelength. The relative exposure factors for the three filters to get similarly exposed images was about 1:4:100 - this compares with 1:2:16 for daylight. So a technically correct WB would result in an image with little or no red. But I wanted a full range of colour, so I white-balanced against fluorescing white paper.

Lighting was also an issue. Painting with a torch means that there will be lighting differences between the three exposures leading to rainbowing in the final image. Using flashguns would have avoided this, but the ZWB1 filters I have for my flashguns leak too much IR.

The first few images are of Lilies. Here's a visible image for reference:



And here's a UVIVF image:



Here's the colour UVIIRF. This was my first try and is not brilliant. But I didn't re-do it because of the time it took (10 minutes for the 1000 nm image). There is also a lot of noise on the 1000 nm image, probably both from long exposure and high ISO):



And for reference, a straightforward reflected IR image (R72 filter, flash):



The next shot is a lily again, UVIIRF colour:



And again a lily. The objective of this shot was to get the leaves rather than the flower:



And now a different flower - an Alstromeria, I think. First, a visible reference shot:



Now a UVIVF image:



And here's the UVIIRF colour image:

[Stefano]

Bernard, that's a beautiful idea! Never thought about this, I would be interested in more.

I like your first attempt the most. The rainbowing is probably an error as you described (uneven light painting), but looks so pretty.

White balancing is especially difficult here. Apart from the different intensities in different bands, how do you make an accurate target? Either you use a broadband IR LED/tungsten lamp, or you have to make something which has a white fluorescence in IR, similar to how the paint discussed here works.

Edited by Stefano, 25 November 2020 - 13:04.

[dabateman]

Clearly you type too much. That keyboard is much cleaner than the top of the desk.
Jokes a side.
The flowers look great. I never finished my attempt at this. Have been too busy. But I have similar filters.

[Bernard Foot]

dabateman, on 25 November 2020 - 12:56, said:

That keyboard is much cleaner than the top of the desk.

There'a no hiding from dustand dirt with UVIVF. I'm shocked by how dirty the carpets are under UV even after they have been vacuum cleaned. And my camera gear turns out to be filthy!

[Andrea B.]

Ah well, we've always lived with dust, germs, dirt that we cannot see. BTW, never shine your UV-LED torch onto a bathroom sink, toilet or floor. It is horrifying. A wise housekeeper once told me that nothing is ever really clean, it simply appears clean.

Very interesting results with the 3 IR-pass filters. This seems to work well enough for IR fluorescence. I have had not interesting results at all for reflected IR using 3 IR bandpass filters. Maybe I'll find time (someday!) to try them for IR fluorescence.

(I also think your 2nd blue photo above is very good. There is a beautiful play of fluorescence and dark areas.)

[Andy Perrin]

I tried reflected IR with three bandpass (two NIR and SWIR) awhile ago and got some color from that. I haven’t tried it with fluorescence though.

[Stefano]

Andy, you may even try UVISWIRF (UV-induced SWIR fluorescence) or NIRISWIRF (Near IR-induced SWIR fluorescence) someday or other kinds of odd things, just make your imagination run. There's so many things you can do with your TriWave.