• Ultraviolet Photography

Pokeweed Berries [multispectral set]

2 replies to this topic

#1 Mark


  • Members(+)
  • 532 posts
  • Location: Massachusetts, USA

Posted 24 September 2017 - 01:27

I found some ripe pokeweed berries recently and figured I try capturing them in a multispectral set, as a sort of follow up to an earlier post of mine (http://www.ultraviol...dpost__p__13712). I also think I realized the problem with a difficult limitation I've struggled with in these sets.

Here's the stem of berries I had setup, hanging out of a glass of water, propped in the middle of a make shift 'aluminum foil box' (to help contain/concentrate illumination, while also limiting stray light contamination). Notice the berry almost in the center of the image; which I nipped with a pair of scissors to expose the flesh/juice inside (also, this is the center of the crops presented below).
Attached Image: 09-17-2017_21-31-36_vis-baader-lantern.jpg

Starting from the shortest wavelength, in the UV image the ripe berries are quite dark and opaque, as is the stem. The foil background works well in this image as it is bright (efficiently reflects UV) (otherwise there wouldn't have been much to look at in this shot!).
Attached Image: 09-17-2017_19-00-59_andrea.jpg

Shifting up into the visible range, the UVIVF image shows bright glowing tips of white and red on the bottoms of the berries.
Attached Image: 09-17-2017_19-09-28_uvivf-baader.jpg

Shooting the same image after cutting open that berry in the center shows some bright red fluorescence inside. Note that this fluorescence looks brighter here than what I saw by eye. This makes me suspect this is very far-red fluorescence - which is hard for us to see, but easy for the modified camera, particularly with the Baader UV/IR cut filter used here, which allows a lot of red light through right up to its cut-off point).
Attached Image: 09-17-2017_21-28-52_uvivf-baader.jpg

Under IR things do in fact change dramatically, as the stem has become bright/reflective and the skin of the berries has become transparent, revealing the seeds inside.
Attached Image: 09-17-2017_19-12-53_r72.jpg

Further into IR, near the upper limits of normal digital cameras (at around 1,000 nm), things look roughly the same, just a bit less 'contrasty'. Also, there is no longer any of that nice false color in the image.
Attached Image: 09-17-2017_19-15-06_eo1k.jpg

This UVIIF image shows the berries to be somewhat gray/non-fluorescing, with some distribution of fluorescence generally around the base of each berry. Note that the green/unripe berries (not shown here) were more IR-fluorescent overall.
Attached Image: 09-17-2017_19-35-08_mteUVIIF-r72-gg420.jpg

Also note how IR bright that berry juice/flesh is inside!
Attached Image: 09-17-2017_21-23-15_r72-uviif_mte.jpg

I tried shooting a UVIIF shot further up into IR (at ~1,000 nm), but it appears the fluorescence doesn't make it that far (half the image here pushed to show very low level signal).
Attached Image: 09-17-2017_19-16-34_eo1k-uviif.jpg

While I'm happy with the overall results of this set, there are a couple things I'll be thinking about to improve my future sets. First, I'm going to try adding some kind of longpass filter, or equivalent, to my IR filter for the UVIIF images - to ensure that pale blue/purple coloring I'm getting is not bleed-through from the UV source. Secondly, and this is the bigger challenge for me, is improving my UVIVF images - as I can't seem to eliminate VIS light from the scene (non-fluorescent VIS I mean). Notice in the UVIVF images how the stem is nearly the same color as it is in the VIS image.

I typically use exposure times as long as 25 seconds for the UVIVF images. Considering this it finally occurred to me that while this long exposure may be necessary to capture faint fluorescence, it also gives the camera a long time to build up VIS bleed through whatever UV source filter I'm using (none of which are perfect).

So I guess this is a limitation (I mentioned above). If I consider the fluorescence the "signal" and the VIS bleed contamination the "noise", then my signal to noise ratio for faintly fluorescing subjects is very low. If I want to minimize the 'noise' component, I must limit the duration of the exposure - which in turn would limit what I can capture in a UVIVF image.

I welcome any suggestions/techniques/tips on how to get around this limitation!

Shooting info:
- Nikon D750 [broadband] + 50 mm Nikon Series E lens + 20 mm extension tube
Lens filters
- UV: AndreaU-MKII
- VIS & UVIVF: EO 425 nm longpass + Baader UV/IR [stack]
- IR & UVIIF: Hoya R72, or EO1K longpass + Xnite830 [stack]
- VIS: White LED bulb
- UV & UVIVF: 4 x 18" T8BL-B + FL-02 glass

- UVIIF: 2 x [MTE-U303 + Hoya U340]
- IR: 40 W incandescent clear glass bulb
- ISO320
- f11
- 0.5-25s, as needed

#2 Andy Perrin


  • Members(+)
  • 2,685 posts
  • Location: United States

Posted 24 September 2017 - 02:20

Mark, try backlighting them also! Dark field...

Maybe with a suitable pen light. I bet the infrared fluorescence will be interesting that way, if you use the fluorescence itself to illuminate the berry innards.

Edited by Andy Perrin, 24 September 2017 - 02:24.

#3 Andrea B.

    Desert Dancer

  • Owner-Administrator
  • 7,541 posts
  • Location: USA

Posted 01 October 2017 - 19:35

First, I'm going to try adding some kind of longpass filter, or equivalent, to my IR filter for the UVIIF images - to ensure that pale blue/purple coloring I'm getting is not bleed-through from the UV source.

An investigation of the Pokeberry pigment is in order, methinks.

It is phytolacanin (= betanin) which is a betalain pigment of the red-purple (red-violet, magenta) betacyanin type, as seen in the visible photo. Now the question becomes, does phytolacanin (or betacyanin) fluoresce? If so, in what range? What excites the fluorescence?

So far I have found this, but I don't have too much time to pursue it today:
Betacyanin fluorescence is observed in the range of 580-660 nm.
But what about chlorophyll (675-680 nm)? Is there any cholorophyll in the Pokeberry stem or green berries?
How do these pigments play a role in what you are seeing in the UVI-Vis and UVI-IR photos?
Do either of these pigments have IR fluorescence?

Another interesting reference in PDF format: Measurement of Beet Root Extract Fluorescence
In this paper, the charts show that although betacyanin fluorescence peaks around 600-610 nm (of course, subject to the particular experimental constraints in that paper), the fluorescence has a fairly long right hand shoulder into the IR region.
Andrea G. Blum
Often found hanging out with flowers & bees.