Jump to content
UltravioletPhotography

Trial UV shots


Recommended Posts

We're almost "there" in terms of sorting out equipment suitable for our project.

 

The biggest issue has been lighting, tried modified vivitar flashes, which work, but we don't think are entirely suitable for use in a museum. The latest lighting to try are Kaiser lighting units, which appear to have been used in past research (The use of digital photography in systematics, McKay, 2013). However once the lighting arrived, we found the bulbs are Osram bulbs that look like they have a peak at 365nm but not as much of the rest of the 300-400nm spectrum as we'd hoped (http://www.osram.com...uctId=ZMP_58963). All the kaiser information we could find suggested they use some kind of proprietary bulbs with more output.

 

However the lighting does work for the 365nm wavelength it seems. Equipment is a modified D7000, 105mm Nikkon UV, Baader U and UV/IR cut filters, and kaiser lighting in a custom made light tent frame. Images are human visible and then UV, with whitebalance adjusted using click-white methods. I could have made them B&W, but kind of like the old warm tones it gives the images!

 

(Note that the bird specimens are old/ex-museum specimens. They weren't collected for the purpose of this study)

 

post-64-0-29811300-1414745408.jpg

 

post-64-0-19330100-1414745426.jpg

 

post-64-0-69214300-1414745435.jpg

 

post-64-0-62034900-1414745385.jpg

 

post-64-0-76192000-1414745391.jpg

 

post-64-0-82790400-1414745396.jpg

 

post-64-0-43107100-1414745403.jpg

 

post-64-0-00671800-1414745415.jpg

 

post-64-0-51823800-1414745423.jpg

Link to comment
Include UV-neutral targets so you can do a proper "w/b". That'll remove the reddish mush now overlaying the images and robbing them of clarity. Have a look at Photoninja to get a tool which easily manages to attain a better and repeatable colour rendition.
Link to comment

In case you haven't already found it, here's a link http://www.ultraviol...eacock-feather/ to some peacock feather shots which seem to indicate what can happen with structural colours in UV.

 

Dave

 

Thanks. One thing I've noticed, in terms of structure, is particularly "glossy" plumage that is "non-UV-reflective" does seem to reflect a bit of UV back. This is most noticeable at the extreme edges of the birds where the body curves around. There aren't any good examples I have posted of this, I'll see if I can later.

 

On the subject of lighting though, I have discovered that Philips produce "sun tanning" bulbs in the same fit as the Osram bulbs...these apparently have more of a 300-400nm spectrum:

 

http://www.scitec.uk..._uva_rev1.0.pdf

Link to comment

Include UV-neutral targets so you can do a proper "w/b". That'll remove the reddish mush now overlaying the images and robbing them of clarity. Have a look at Photoninja to get a tool which easily manages to attain a better and repeatable colour rendition.

 

I have a colour checker passport and photoninja, just haven't gotten around to figuring out how to use them to their full advantage. PhotoNinjas menu system seems to be contrary to what I am used to! We have some lambertian targets on order too.

Link to comment

I think I've sussed what I was having issues with, the toolbars had become "unpinned". I'd been trying to figure out how to perform all the corrections as I couldn't find them in the top menu bar. Moved the mouse to the side and hey presto, they appeared!

 

Below is a shot with the colour checked. Left side images are as adjusted by photoshop. Right side photos are as adjusted by Photoninja (top right has set the colour profile to the colourchecker, bottom right has had WB set to the top right black square)

 

post-64-0-15149900-1414752369.jpg

 

It does seem the PhotoNinja processing brings out some additional detail in the darker areas. Thanks for the nudge to get me to try it out!

Link to comment

You're on the right track. If you have access to Lambertian targets with a very flat reflectance across the UV band, try UV "w/b" on these as well. The last two b/w tiles on the Color Passport are OK, but you reduce calibration error by using a larger target and average the results.

 

How much to lift the middle tones in a UV capture is a matter of practicality, subject, and personal taste. No fast and fixed rules apply here. However in order to achieve consistency across a series of captures, one should probably don't overdo this last finishing step. UV is in terms of tonality very different from our usual visible world. If contrast in UV becomes excessive, it is often better to work with the lighting setup instead.

Link to comment

You are making excellent progress!

 

Why, may I ask, was it decided that the modified Vivitar flashes were unsuitable for use? Fluorescent UV lamps, while perhaps useful to check focus UV in live view, will not emit a spectrum that reasonably models solar UV & Vis spectral power distribution. Solar simulation is best accomplished with a Schott WG-320 filtered Xe-arc lamp. Next best alternative to Xe is a similarly filtered quartz enveloped tungsten filament lamp. Both Xe and quartz tungsten emit a great deal of visible light and heat. However, both also will have high color rendering index.

 

The UV-neutral targets Bjørn recommends may be sourced in 1.25" and larger sizes from Labsphere or Avian Technologies but you need not pay extra for calibrated kits or singles. I have the 1.25" Labsphere Spectralon® in 99% and 5%. The 99% is easy to "blow" (overexpose) but I find that somewhat useful sometimes to aid me in limiting over exposures.

 

You may wish to review the following postings if you have not already, I found most instructive: Discussion of Profiling in Photo Ninja for UV Photos, Photo Ninja: Colour Correction: Light Source Tab and How to make UV colours reproducible.

 

Link to comment

Thanks John. We felt the flashes weren't suitable due to the environment we're working in. There's a small team of us working around different museum collections. A lot of curators are concerned over use of photographic flashes as it is, so having powerful flashes that also emit UV might be a tricky one to obtain permission for. in addition, constant flashes going off could be distracting for other collections users!

 

Do you have any recommendations on a typical XE-Arc lamp set up? Like the flashes these sound like they might be "too much". A lot of the specimens we plan to work with are over 100 years old so we want to be careful not to frazzle them!

 

We have labsphere targets on order, but there seems to be a delay somewhere along the lines ;)

Link to comment

No, I cannot recommend a Xe-arc source that would likely be suitable given that concern. The Vivirar flash is a Xe flash and if I were you I would propose to the curators to use it fitted with additional filtering. You may recall I described this 02Oct14 to your query regarding UV Lighting and health.

 

Adding a short wavelength UV-B blocking filter would not only reduce photobiological risk to you and your colleagues but it would remove the most energetic UV radiation that could degrade specimens. The Badder U-filter you will be using is specified as having a 60nm HBW (Half Band Width) 320-380nm. This notation usually means that the filter has 50% transmission (Half) at 320nm and 360nm. A Schott WG-320 long pass filter also has 50% transmission at 320nm. Placing a WG320 type filter on the flash will not significantly degrade the output relative to the transmission band of the U-filter. You would then also add some type of black glass UV bandpass filter to further attenuate visible light to protect artifacts sensitive to visible & NIR wavelengths. This is the typical type of dark filter used for UV-induced visible fluorescence photography. This removes most of the long wavelength radiation that is not going to be transmitted by the Baader U on the lens anyway, so if you are not using it you don't need it. The point is that this combination of filters also protects sensitive specimens from needless exposure to UV-B, visible and NIR to a level that is as low as reasonably achievable, aka ALARA, a tenant of radiation safety practice, while still enabling the research.

Link to comment

I forgot a couple of points I wanted to make above.

 

A continuous UV source is easily more hazardous/damaging because exposure accumulates. A low power UV lamp running for ten times longer is the same exposure as a UV lamp ten times more intense running for 1/10th the time.

 

Your visible reference photos should use whatever sources the curators normally use for display purposes assuming good color rendering is achievable.

Link to comment

I concur with John that you definitely want the flash over the steady lamps in order not to "fry" your specimens with heat and excess UV at close range. However if working around other folks, you do need to contain the flash within a shooting tent or somesuch. That is easily accomplished by setting up some fiberboard or even cardboard panels around your shooting area.

 

Remembering that Birds have UV vision and are constantly in the sunlight when alive, I can't imagine that a couple of UV flashes are going to harm dead specimens. Their feather makeup is meant to withstand more than a little UV.

 

**********

 

My current protocol with the CC Passport and Photo Ninja is to photograph the CC Passport AND a white standard in UV and in Visible light at the beginning of each shoot using the respective UV or Vis illumination sources for that shoot. I occasionally rephotograph the CC Passport and white standard during a shoot if any ambient lighting changes occur which could affect the colours.

 

Later I create for that shoot a Custom Visible Profile in Photo Ninja. After I have created the Custom Profile for the shoot, it is easy to store the profiles and a few meaningful settings in a Photo Ninja Preset for batch application to the Visible shoot photos. Further edits may be necessary, of course. But it is very work-saving to use Presets.

 

Added: For the UV, I create a White Balance Preset (not Profile) in Photo Ninja as you cannot create an actual Profile from the CC Passport. Again the Preset is great for batch application to the UV shoot photos. So the white standard is required for a UV profile, while the CC Passport is optional. However the CC Passport is useful anyway because after the white balance step using the white standard, the patches on the UV photograph of the CC Passport "turn blue". So it is a kind of sanity check to UV-photograph the CCP along with the white standard. "-)

 

Photo Ninja Custom Light Profile: http://www.ultraviol...__fromsearch__1

 

I can suggest some typical settings to include in the profile Presets, if you like.

Especially for the UV Preset as I have not (I think?) addressed that anywhere.

Link to comment

I wanted to add that using broad spectrum UV lighting will give a larger range of false colour in the UV photographs. If you use an illumination source which is narrower and peaks around 365nm, you will tend to get a more monochrome output. I am thinking that it would be best -- in the sense of being "truer to nature" and also "more interesting" -- to photograph the birds in such a way that the UV outcome is more like what the birds would actually see in real life - although to them whatever colours they perceive are not false of course. ;) And certainly not the same as the editor false colours we produce.

 

I had a point there when I started. I wonder if I made it??

Link to comment

Dang-rabbity folks, those UV bird photographs above are REALLY interesting. Wish I was doing this project !!!

 

Found this:

Ultraviolet (UV) light perception by birds: a review

by J. Rajchard, Faculty of Agriculture, University of South Bohemia, Ceske Budejovice, Czech Republic

http://www.vri.cz/docs/vetmed/54-8-351.pdf

Link to comment

Remembering that Birds have UV vision and are constantly in the sunlight when alive, I can't imagine that a couple of UV flashes are going to harm dead specimens. Their feather makeup is meant to withstand more than a little UV.

 

Excellent point Andrea!

 

However, what birds or any other wild creature were never exposed to is UV below the solar ozone/atmospheric spectral cut-off. For environmentally relevant UV imaging extraterrestrial wavelengths are rather a like a contaminant in a chemical reagent. By filtering to attenuate these wavelengths a UV-flash is much less likely to injure live critters - including us too. ;)

Link to comment

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...