Why is this? (2)

[bvf]

I photographed some forget-me-nots this evening. As usual I took visible & UV spectrum comparison shots.

 

Here is the visible shot:

 

And here is the UV shot:

 

Why are the three flowers in the UV shot of such different shades of blue when in visible they are all the same colour?

 

The UV shot was taken through U340+S8612 filters, white balancing against PTFE, with 3 flashguns (with UV-absorving lenses removed) very close to the flowers.

 

The visible shot used a ringflash.

 

So the lighting was quite different, but I don't think that is the cause. If the angle of the lighting was relevant, then there would be variations across each flower, and the overlapping petals would be the same as the petal behind.

 

I notice on the visible shot that the central ring of one of the flowers is yellow whereas it is white on the other two. So maybe the flowers simply have different colouring which only is evident in the UV.

 

Any other thoughts on this?

[Andy Perrin]

Why are the three flowers in the UV shot of such different shades of blue when in visible they are all the same colour?

Ok, I think there is a misconception here that each visible color maps to a fixed UV false color. That is not the case.

 

I notice on the visible shot that the central ring of one of the flowers is yellow whereas it is white on the other two. So maybe the flowers simply have different colouring which only is evident in the UV.

Yes, I think that's it.

[nfoto]

The Myosotis flowers (and other members of the Boraginaceae) typically show a change in colour during their development. This change besides being very evident in UV, is also seen in visible light. Perhaps they simply reflect altered chemcical stuff in the floral parts over time?

[nfoto]

The flower structure seen in the Boraginaceae is a cyme which typically unfurls from top in the same manner as seen in ferns. Interestingly, the difference in UV appearance holds up even when the flowers are all white (albino form). The cyme below belongs to Myosotis sylvatica, which often is seen in albino form.

 

UV

 

 

Visible

 

 

(captures with Panasonic GH-2 and UV-Nikkor; SB-140 flash and Baader-U for UV, daylight and Baader UV/IR cut for visible light)

[DaveO]

Hello Bernard,

 

Here is a link to a previous post to the question of false colours

https://www.ultravioletphotography.com/content/index.php/topic/3040-for-the-botanists-flowers-response-to-uv-and-lack-of-it/

 

Those of us who have spent too much time wondering about the apparent correlation between visible and false-UV colours sometimes can almost see an answer but then it slips away like the morning dew.

 

Dave

[bvf]

Thanks for that input, Andy, Dave & Birna.

 

I followed that link, Dave, and have picked up the paper on UV pigmentation in flowers that you linked to. I'll have a read of that.

 

On correlation of visible and false-UV colours, I have been using the following link, which I'm sure you're all familiar with.

 

http://www.savazzi.net/photography/uv-b.html

 

Bernard.

[Andy Perrin]

Yes, that Savazzi link is fine as far as it goes. But while certain wavelengths map to certain colors, it does NOT work in reverse (or only crudely).

[dabateman]

Bernard,

The discussion of color in the various portions of the UV spectrum has been discussed here many times. I am not sure what camera was used in Savazzi tests. But Jonathan has tested a Canon camera and I have experience with my Olympus camera with a Panasonic sensor.

In my hands 300nm is very very green, most of the response is only in the green channels of the sensor. When I see blue coming back in, at wavelengths less than 380nm its IR leakage.

300nm very very green

313nm very green

335nm green

370nm yellow

380nm blue

390nm violet/blue

405nm blue and loss of typical UV dark patterns.

 

But I haven't tested a Sony sensor and there may be slightly different dyes used on them for the three color channels. However, from my research it seems most sensor companies buy the dyes from Fuji. Which I was surprised by. So they may be more similar than we realize.

 

Back to your topic. I have seen different flower colors due to development. So my gut reaction would be your 3 flowers are different ages. If you get a live plant could be fun to do a uv stop motion animation. Look at the color change over time, from birth to death.

[bvf]

Thanks, Andy and Da.

 

I've just switched from a Canon EOS M to the Sony a6000 for full spectrum. I did a comparison shot between the two - see below. They are pretty similar: the Canon colouring is slightly more violet tather than blue - may be because I exposed it at ISO 400 whereas the Sony was at ISO 200 (because from the LiveView images it looked as though the Sony was more sensitive, which I now doubt).

 

I'm never going to see the greens in your listing as with my combination of lenses (Cassar S or El-Nikkor 105mm) and filter (U340+S8612) I struggle to get to 320nm. May change when my Baader U finally arrives (how can it take 3 weeks for Amazon to ship from Europe to the UK ???). The Savazzi data includes a "gray/rust" region, and I certainly get a lot of grey in my images.

 

This was the shot (Mexican Orange Blossom) taken on the Canon:

 

And this is on the Sony:

 

Bernard

[nfoto]

Beware of what profiles your converter applies. For example, Photo Ninja defaults to "daylight/flash", but this assumes we are working with visible light. Better to use "No profile" instead. Typically this will lead to more blue/cyan hues than the purple ones.

[dabateman]

Don't expect much less than 330nm withthe BaaderU. Mine cut off at 325nm. But has 70% transmission at 330nm. Here is my transmission curve of my filter when I still worked in a lab:

https://www.ultravioletphotography.com/content/index.php/topic/2665-sigma-dust-block-filters-bw486-and-my-venus-2-spectra/

Its the last spectrum.

However, the contrast with the Baader venus filter, I like the best. Very nice and still my go to UVa filter.

To go into the UVb area you would need a quartz lens. My Nikkor 80mm f5.6 enlarger lens can just see at 313nm. Buts its a horrible noisey image.

But too be honest, you may not want too fall into the Uvb rabit hole. Its very very expensive with lens options and filters. Then your always looking for best lights. I finally got a 302nm bulb coming in the mail. Was very cheap, but will have to figure out how to drive it. I don't yet know what it is, just hope it falls into some standard.

 

[bvf]

Da - thanks., Yes, my ambitions don't extend to UVB, and I'm not up for the cost of a specialised UV lens, so I'm trying to get the best I can in the UVA region with off-the-shelf kit. And I'm enjoying it!

 

Birna - OK, you've lost me! I'm OK with white balancing, but don't understand "what profiles your converter applies" or how to use "No profile". Sounds like this is a camera setting, but I've never seen it. How do you go about setting this profile. (I'm sure there's info. about this somewhere in UVP!)

 

Bernard.

[Andy Perrin]

Bernard, are you shooting RAW?

[bvf]

For most photography I shoot RAW + JPEG - but only use the RAW for a few things like re-white balancing or lens correction. But for UV flower stuff, JPEG only. Reason is that I always use focus-stacking, and may have 100 shots to process (200 for a stereo pair). The focus stacking software can't take RAW. I could process all 200 images in RAW first, but to be honest my patience won't stretch that far.

 

From your question I assume that profiles are a RAW thing. Looking at RawTherapee, the RAW software I use, I see there is a section on Colour* Management > Input Profile with a "No Profile" option. Guess that's it. I'll have a play with that, although it won't help when focus stacking.

 

* Yes, with "colour" spelt correctly!

[Andy Perrin]

If all 200 images are being edited identically (which is probably what you want anyway?) then you can do a batch edit. PhotoNinja can certainly do this. I am not sure about RawTherapee.

[nfoto]

RawTherapee has batching capabilities too.

[bvf]

Yes, I had a look at rawTherapee yesterday and found how to do it. Being able to process the RAWs could be a real boon - the central areas of flowers in UV are often so dark that you lose detail, and adjusting the brightness and contrast in the final JPEG is sub-optimal. I'll have a go at pre-processing the RAWs before I go to the focus-stacking stage.

[nfoto]

Dump the processed TIF files to a project folder on scratch disk, then run the stacker software on that folder. When you are satisfied with the outcome, just delete these files. Do keep the RAW files though.

 

I use dedicated machine(s) for stacking so I can easily multitask during the often time-consuming stacking process. I run the RAW -> TIF conversion on these systems as well, using either Photo Ninja or (in special cases) RawTherapee.

[bvf]

Is there a noticeable advantage to using TIF? I've always used JPEG, but for no particular reason. (I assume TawTherapee can output TIFs and that Zerene can handle them.)

 

I agree that it is a pain waiting for the stacking software to run. I tend to multi-task on my single desktop machine - while one set of images is stacking I'm selecting the next set for stacking, or Photoshopping/GIMPing the stacks that have already been created, or creating the stereo image from the already stacked left and right images. So generally I can avoid getting bored while I'm waiting.

[nfoto]

In order to keep the maximum of detail, 16-bit TIF is superior to 8-bit jpg. This is in particular seen when the output file needs further post-processing.