• Ultraviolet Photography
  •  

Botanical z-stacking in UV

10 replies to this topic

#1 Daniel Geiger

    Daniel Geiger

  • Members
  • 9 posts
  • Location: Los Angeles, California

Posted 02 January 2017 - 17:43

Greetings,

I am curator of malacology (study of mollusks) at the Santa Barbara Museum of Natural History in California, where I am also in charge of the scanning electron microscopy facility. I have a secondary interest in systematics and evolution of a tropical orchid genus (Oberonia) and am a Visiting Research Scholar with the Huntington Botanical Garden - Botany in San Marino (Los Angeles), California. For the latter, I am interested in fine structure of minute 1-5 mm flowers, and given the different cell surface sculptures, I wonder whether there are any UV patterns for potential pollinators. I have been fiddling with this for a couple of months now. I have gotten UV z-stacking to work; plan on posting a method later.

UV equipment currently is a Canon 5D mark II full spectrum, Nikkor EL 80/5.6 (old, not N), Canon MP-E 65 mm (works, but 10x lower transmission than Nikkor), 2" Baader U Venus, 5W 365 nm LED flashlight.

Questions currently include:
- Spectral distribution of LED UV sources? Too narrow to make false color images possible?
- Looking into modifying flash for UV flash. As far as I understand, old-cheap is better for UV. Any particularly good/cheap ones? PC triggering no problem. Should withstand repeat triggering (20-100 shots, typically) at 1-2s interval, reasonable consistency in light output. Should I add a UG11 in front of flash, or is Baader U on lens enough?
- False color UV: possible with Baader U or do I need some visible light component (UG 5-type filters)?
- Channel swapping vs channel mixing: the latter seems to do the same thing, but much easier. Am I missing something?
- Anybody tried a Paul Buff Einstein 640 for UV? Possible?

Other imaging equipment is varied:
ArcaSwiss 4x5 kit from 72 - 720 mm
Canon 5DsR mainly with Zeiss primes, Cognisys Stackshot
Zeiss Discovery V20 mot Stereomicroscope with Axiocam HRc, a couple of planapos.
Zeiss Axioskop2plus compound microscope, motorized with Cognysis stepping motor, set of planapos, one epiplan neofluar
At work: Zeiss EVO 40 XVP SEM.

Photography background:
SLR for 35 years.
A bunch of classes in scientific photography in dept phys chem University of Basel, Switzerland.
A few photography publications:
- Geiger, D. L. 2006. Chapter 7. Applied film photography in systematic malacology. In: The Mollusks, A Guide to their Study, Collection, and Preservation, Universal Publishers, Inc., Boca Raton, Florida. C. F. Sturm, T. A. Pearce and A. Valdes (eds.). 73–100.
- Geiger, D. L. 2013. Imaging small orchid flowers using visible light. Orchid Digest 77: 112–123.
- Geiger. D. L. 2011. Taking pictures of tiny orchid flowers. American Orchid Society website http://www.aos.org/Default.aspx?id=355
- Geiger, D. L. 2013. Considerations and limits of z-stacking in macrophotography. PhotoTechniques Sept/Oct. 2013: 34–36.
- Represented by alamy.com stock-photograph agency [though not happy with them anymore]
- Partial list of published photographs: http://www.vetigastr...ions/photos.php


#2 Andy Perrin

    Member

  • Members(+)
  • 818 posts
  • Location: United States

Posted 03 January 2017 - 01:29

The Baader U does make UV false color pictures just fine, provided 1) that you have a wide-band source (not LED flashlight) and 2) that you do a UV white balance as described in our stickies. The majority of the pics on this site were made under those conditions.

The spectral distribution of 365nm LED flashlights is too narrow to get a really good false color image in my experience? You may be able to rig something up with different LEDs at a variety of wavelengths. I just use sunshine for everything. I think most people here use a flashlight for focusing though.

Edited by Andy Perrin, 03 January 2017 - 01:31.


#3 Andrea B.

    Desert Dancer

  • Owner-Administrator
  • 5,094 posts
  • Location: USA

Posted 03 January 2017 - 02:25

Hello Daniel and Welcome to UVP!
I hope we can answer some of your questions. There is a broad range of experience here on our quiet little website. I'm looking forward to seeing some of your work on tiny flowers, so do keep us updated.


I'll write a few answers and ask other members to please chime in.

Spectral distribution of LED UV sources? Too narrow to make false color images possible?
My experience with Nichia-chipped UV-LEDs having peaks at 365 nm and 385 nm has shown that false colour is slightly more difficult to obtain, but certainly not missing entirely under UV-LED light. Some converter/editors are better than others at bringing it out.

The bandwidth is fairly narrow with UV-LEDs. Using a UV-flash provides a broader spectrum light with more false colour possibilities. I don't have any charts at hand for the Nichias, but they can be found on the manufacturer's websites. We mostly use UV-LEDs for inducing visible fluorescence and use UV-flash or Sun (or a combo) for reflected UV photography.
Here is a link to photographs of an Erigeron philadelphicus showing UV photos made in Sunlight, under a 365 nm UV-LED and a 385 nm UV-LED. The latter two photos exhibit a false blue that is somewhat duller than the image made in sunlight. However, given that this blue is false, we are free to push the satuation slider if desired for artistic purposes.
http://www.ultraviol...__fromsearch__1

Looking into modifying flash for UV flash.
Here are two links for further reading about UV-flashes and flash modification.
  • The second is a search tag which will bring up a lists of posts about UV-lighting and UV-flashes.
    UV Lighting
Should I add a UG11 in front of flash, or is BaaderU on lens enough?
You would have to know whether your particular flash outputs enough Visible or IR light to somehow affect the botanical specimen which is being photographed. And also note that the BaaderU passes a very small amount of violet light which might also be of concern. Remember, the border between UV and Violet is "somewhere between" 380-400 nm, so draw the line where it suits. I think most of us have some kind of filtration on our UV-flashes. But, for example, my UV-flash still outputs a bit of Vis/IR and that -- so far -- has not affected my UV botanical work under the BaaderU or other UV-pass filters like the StraightEdgeU or a U360+S8612 filter stack.If the UV-flash or UV-LED is used for inducing visible fluorescence photography, then filtration on the light is mandatory to ensure only UV-illumination.

False color UV: possible with Baader U?
Yes, see any of our botanical sections, listed below. All UV images are rendered in a standardized UV false colour palette of blue/yellow/grey/black/white which quite naturally occurs using a click-white tool during raw file conversion. Actually you can look in any of the non-botanical sections also. :D
It should be noted that many botanical specimens do not exhibit much false colour at all even under broadband filtration and broadband illumination. You will find quite a range amongst our posted examples.
UV Wildflowers by Family
UV Cultivars: Garden & Decorative Flora
UV Cultivars: Vegetables, Herbs & Crops
UV Other Botanicals


Channel swapping vs channel mixing: the latter seems to do the same thing, but much easier. Am I missing something?
I don't know what you are attempting to do, so don't know how to answer the question.
Last time I swapped a red channel with a blue one, I used a channel mixer to do that.
((But I don't usually use channel swaps in my own UV or IR work.))

Anybody tried a Paul Buff Einstein 640 for UV?
Not that I know of. However, Bjørn Rørslett uses a Broncolor studio light for reflected UV photography. There is a link in the Sticky referenced above. A key point to consider when using studio lighting is that it is heavy and that it can be very hot on small botanical specimens.
Possible?
As currently listed on the Buff website, NO. Because the Einstein 640 flashtube is coated to block UV and the lamp dome is also coated. However, you could contact the Buff team and ask if they have uncoated flashtubes and domes available.
Andrea G. Blum
Often found hanging out with flowers & bees.

#4 Andrea B.

    Desert Dancer

  • Owner-Administrator
  • 5,094 posts
  • Location: USA

Posted 03 January 2017 - 02:38

For the latter, I am interested in fine structure of minute 1-5 mm flowers, and given the different cell surface sculptures, I wonder whether there are any UV patterns for potential pollinators?

Yes.

Here is a little 6mm Galium tinctorium flower with a bit of two-tone patterning on its tiny petals. The tips are slightly darker than the centers. Conical cells are evident on the surface.
Galium tinctorium [Stiff Three-petaled Bedstraw]
Attached Image: galium.jpg

This Myosotis stricta is about 3mm across and shows a dark bullseye center in UV. UV appearance changes with age in many Myosotis.
Myosotis stricta [Strict Forget-me-not]
Attached Image: myo.jpg

And one more - Veronica arvensis, another tiny 3mm flower with a bullseye and striping on the petals. Veronica arvensis [Corn Speedwell]
Attached Image: veron.jpg
Andrea G. Blum
Often found hanging out with flowers & bees.

#5 Andy Perrin

    Member

  • Members(+)
  • 818 posts
  • Location: United States

Posted 03 January 2017 - 03:00

I should add that my LED flashlight is NOT a Nichia, so they may behave a bit differently. Mine is some generic brand that I can't recall offhand. I could check when I get home.

Edit:Tuofeng 1W 365nm

Incidentally a spectrum is shown in that Amazon listing. I have no idea who made it, why it is shaded the way it is[*], or anything else about it, but...it's consistent with my experiences anyway? Certainly mine seems deficient in the shorter wavelengths.
Attached Image: 41+3R1I-yjL.jpg

[*] It's possible that someone is defining the cut-off for UV to be 380 and under.

Edited by Andy Perrin, 03 January 2017 - 20:46.


#6 enricosavazzi

    Member

  • Members
  • 374 posts
  • Location: Stockholm, Sweden

Posted 04 January 2017 - 13:12

I use almost always one of the following Bowens studio flash units for studio UV imaging, and I am satisfied with them. I should stress that you need to replace the tubes (and the dome in the 1500 Pro) with uncoated ones, sold as accessories by Bowens.

http://savazzi.net/p...ens1500pro.html
http://savazzi.net/p...obes.htm#Bowens

Measured with a cheap Sekonic L-308S flash meter, output of the two Bowens units, at all powers I measured, is within 1/4 (edit: 0.1) stop on any number of successive flashes. The 1/4 (edit: 0.1) stop deviations might be caused by the meter and/or flash, I cannot be sure. I don't see any noticeable changes in exposure in successive images recorded by the camera, so I suspect the meter cannot measure repeatably with a precision of 1/4 (edit: 0.1) stop or lower. Recycle times at the power levels I normally use are under 1-2 s. I have had consistency issues with power output, instead, with a cheap studio strobe from China (fluctuations of over 1 stop for no apparent reason).

As mentioned earlier by Andrea in this thread, studio strobes can easily damage botanical specimens by overheating them, unless used at really low power. I have noticed that some potted plants tend to wilt a few days after a flash photography session, even when no damage is immediately obvious. It might be a good idea to expose only the area of interest to the radiation, and protect the rest of the plant, perhaps with a shield made from aluminium foil between flash and subject, with an opening large enough for the subject area (I have not tried this, so I don't know if it helps in the long term or if the exposed area will wilt anyway).

For small subjects, it should be feasible to mount a NIR-cut filter between flash and subject, which would remove roughly half the radiation output of the flash. A VIS- and NIR-cut filter would be even better to absorb most of the energy, but then you have additional problems:

- Large UV-pass, VIS- and NIR-cut filters are very expensive, if available at all
- The filter may overheat and crack
- You must remove the filter to record VIS images, in addition to swapping the filter on the camera lens

I use an MTE UV torch for framing and focusing in live view, or sometimes a battery of 8 UV fluorescent tubes for larger areas, but not usually as a light source for shooting.

So far, I have not found practical, truly broadband continuous UV sources for shooting. I once bough a continuous xenon-arc source for paint hardening, but its UV output was disappointingly low for a device twice as large and heavy as a tabletop computer. Its UV output at the attachment port for a light guide, used to illuminate a small subject and recorded by full-spectrum converted cameras, was no higher than an MTE 3W UV torch at the same distance. The xenon-arc bulb was new, so it was not a matter of a spent bulb in need of replacement.

Edited by enricosavazzi, 04 January 2017 - 14:56.

-- Enrico Savazzi

#7 Daniel Geiger

    Daniel Geiger

  • Members
  • 9 posts
  • Location: Los Angeles, California

Posted 05 January 2017 - 05:37

Thanks for all the replies and links. I'm definitely in the right spot here! Also had a great off-line exchange with Klaus Schmitt.

I realize I have to convert a couple of flashes. I think the Vivitar 283 will be sufficient; don't need zoom head. Re studio flash, was mainly wondering whether I can use what I already have. Converting Vivitars seems more straight forward than messing with the flash tubes on a studio unit.

I have used MT 24 EX for visible light, and no problems with the flowers; confirmed the lack of problems with SEMs. I assume some flowers are hardier than others. I seem to be lucky in that regard. I have noticed that at low power output, consistency decreases, but at most maybe 1/3 f-stop. [1 f-stop on studio strobe is a bit much. The PCB Einstein is also within 0.1 f-stops as measured with a Sekonik 558 flash meter, and not sure whether accuracy of the flash or the meter is off.] Usually I set the MT24 to 1/64 power (lowest available), even then have to add ND filters to get it in range. My main visible light set-up is with a MP-E 65 at f/2.8 and a 5DsR at ISO 100.

Re violet leak, that is of no concern. My main interest is "some pattern with UV" or "no pattern with UV" compared to visible light. False color will be helpful for communication purpose, but not a goal in itself. By using a broad range UV source, I can also cover as much territory as possible. The final product will be a couple of pages in a botanical monograph.

I think channel mixing can achieve same as swapping in single frame/spectrum exposure. Only in multispectral composite imaging, you cannot get around channel swapping.

off to Ebay ...

#8 Andrea B.

    Desert Dancer

  • Owner-Administrator
  • 5,094 posts
  • Location: USA

Posted 06 January 2017 - 02:54

Something like the Blak Ray might also be useful?
I'll be back with the links in a minute. (Gotta switch to my laptop from iPad.)

http://www.ultraviol...__fromsearch__1

Like Enrico mentioned, I have also experienced plant demise a few days *after* the UV photo session. Let's just say that if you have a valuable specimen, be aware you might be putting it at risk with 40 flashes.
Andrea G. Blum
Often found hanging out with flowers & bees.

#9 Daniel Geiger

    Daniel Geiger

  • Members
  • 9 posts
  • Location: Los Angeles, California

Posted 06 January 2017 - 08:50

Thanks for the pointer. Will first try flash. I'll keep the potential damage in mind.

On an other note, I how do I get e-mail notifications if someone replied in my posts? I "follow" the topic, and set notification options to "auto follow topics I replied to" "immediately" and notification method to e-mail and notification list. Somehow I don't get any email notifications at all and notification list is empty. I do get such e-mails from other fora.

#10 Andrea B.

    Desert Dancer

  • Owner-Administrator
  • 5,094 posts
  • Location: USA

Posted 07 January 2017 - 07:12

Daniel, it looks like all your notification settings are correct. So I'm not sure why you have not gotten any emails. Did you verify your email address was correctly listed? Tomorrow I will log onto our server and check for email output errors and let you know what I find.
Andrea G. Blum
Often found hanging out with flowers & bees.

#11 Andrea B.

    Desert Dancer

  • Owner-Administrator
  • 5,094 posts
  • Location: USA

Posted 09 January 2017 - 17:30

Daniel, I found no email errors in our logs. So I am not sure why you have not received UVP notifications. Please check your spam filters to see if they are blocking the notifications.
Andrea G. Blum
Often found hanging out with flowers & bees.