• Ultraviolet Photography
  •  

Disocactus ackermanii [Orchid Cactus]


30 replies to this topic

#1 msubees

    Zach Huang

  • Members
  • 389 posts
  • Location: Michigan, USA

Posted 11 June 2014 - 21:54

Huang, Z.Y. (2014) Disocactus ackermannii (Haw.) Barthlott (Cactaceae) Orchid Cactus. Flowers photographed in ultraviolet and visible light. http://www.ultraviol...-orchid-cactus/

Okemos, Michigan, USA
11 June, 2014
Cultivar in home garden

Synonyms:
  • Epiphyllum ackermannii Haw.
  • Cactus ackermannnii Haw.
  • Cereus ackermannii Haw.
  • Phyllocactus ackermannii Haw.
  • Nopalxochia ackermannii Haw.
  • Phyllocactus weingartii A. Berger
  • Nopalxochia conzattianum T.MacDoug.
  • Pseudonopalxochia conzattianum T.MacDoug.
  • Nopalxochia ackermannii var. conzattianum T.MacDoug.
  • Disocactus ackermannii var. conzattianum T.MacDoug.
Common Names:
  • 令箭荷花 lingjian hehua (Chinese)
Comment:
Photographs taken during a rainy day around 9 am. Flower petals show some UV reflection but it is rather uniform and do not show patterns. Young anthers are darker under UV (purplish in false color) while dehiscing anthers are lighter (light blue in false color).

Reference:
1. Wikipedia (11 June 2014) Disocactus ackermanii. Wikimedia Foundation, San Francisco, CA.


Equipment [Panasonic G5-broadband + El Nikkor 80mm f/4.5]

Visible Light [f/8 for 1/8" @ISO160, S8612 (1.75 mm thick), natural light]
Attached Image: P1090990-disocactus-vis.jpg

Ultraviolet Light [f/8 for 15" @ISO160, S8612 (1.75 mm thick) and UG 11 (1 mm thick), natural light]
Attached Image: P1090993-disocactus-UV.jpg

Edited by Andrea B., 20 July 2014 - 15:12.


#2 msubees

    Zach Huang

  • Members
  • 389 posts
  • Location: Michigan, USA

Posted 11 June 2014 - 21:56

well, my virgin botanic post...not sure if WB is right. got my virgin teflon today so I can shoot that one and correct for this. But silkpix does not seem to have point-selection for WB.

#3 Andrea B.

    Desert Dancer

  • Owner-Administrator
  • 5,066 posts
  • Location: USA

Posted 12 June 2014 - 02:54

Well, Zach, perhaps a reshoot would be in order. We are looking for a blue/yellow false colour representation in our UV presentations, not a magenta and pale red/orange look. Working with your Teflon should help fix this.

I'm more concerned about the possibility of your UV filter stack leaking some IR contamination? The flower you are showing us is not uniformly UV dark as you write. Indeed, it would appear to be fairly UV reflective. Now, I do not know what Orchid Cactus flowers look like under UV, but I have seen a lot of other very, very UV dark cactus flowers. Maybe Disocactus is an exception to this, but maybe we should also consider whether your 1mm S8612 is thick enough to suppress all the UG11 IR leakage ?? If you are confident there is no IR leakage, then please ignore this paragraph.

For the Visible shot, it would be useful to dial back the red saturation a bit by shooting with a Neutral or "Natural" in-camera colour setting. (I forget jusst now what this is called in a Panasonic.) Also the white anthers are overexposed and have lost all detail. Sometimes a little fill flash with a higher shutter speed actually helps bring out this detail. So experiment a bit more with this Visible red flower by trying some alternate exposures. Vivid red is always a tough colour to shoot, and it seems like all digital cameras have a tendency to blow the reds. I have a lot of trouble with red flowers myself - and sometimes purple ones too.

You should add the botanical authors to the data: For Disocactus x hybridus this is (Van Geel) Barthlott.

Please don't give up !! :lol: We are looking forward to lots of contributions from you and from all our other members. ;)
Andrea G. Blum
Often found hanging out with flowers & bees.

#4 igoriginal

    Iggy

  • Members(+)
  • 582 posts
  • Location: Jackson, Mississippi

Posted 12 June 2014 - 05:28

@ Zach,

I would have to agree with Andrea. When I first saw the UV image, earlier today when you first posted it (a quick glance-over) ... I had the impression that there is some IR contamination, here.

For one thing: I see purple (anthers), in places where I expect to see black or nearly black (UV-absorptive parts).

I also expect to see the typically grayish-silver background of a UV-neutral white-balanced image, via PTFE. Not the case, here.

And, I suspect that the culprit is that your IR-block glass in too thin. I, personally, would never use S8612 cut any thinner than 1.75mm. Speaking from my own (but limited) experiences with IR contamination (and I do have various thickness combinations which I have gone through, of which all of my 1mm thick IR-block filters are now sitting in a shelf somewhere, lonely from abandonment).

All of my own stacks are fitted with a minimum S8612 thickness of 1.75mm (although I personally feel most safest with 2mm thick, even if it's a bit overkill, admittedly).

Anyway, I could be wrong (and there is a chance that I am), but I truly got the initial impression that some IR contamination is the culprit, here.

(Not to mention: If your IR-block glass is too thin, it would also explain your fruitless and endless struggle to set your UV colors to "UV neutral", via PTFE. With IR contamination present, obtaining a sufficiently UV-neutral-balanced image via PTFE is just about next to impossible, from my own experiences.)

Also, aside from going with the option of setting your camera to a more "de-saturated" mode to not blow out the reds (in VIS images), I find that an even easier (and faster) solution is to slightly (and deliberately) under-expose the VIS image. This works wonders for keeping reds (and also yellows) in line, because with deliberate (but carefully reserved) under-exposure, the bright reds and yellows are usually the first to become reduced in saturation and / or brightness, without affecting other color saturations at the same rate of reduction.

It has always worked for me, anyway, to control red and yellow over-saturation and / or over-brightness. (You can always adjust highlight / shadow curves later, to compensate for other areas).

But, otherwise, you seem to be on the right track. Great to have you jump into the game!

Edited by igoriginal, 12 June 2014 - 06:06.

Igor Butorsky

#5 Andrea B.

    Desert Dancer

  • Owner-Administrator
  • 5,066 posts
  • Location: USA

Posted 12 June 2014 - 12:47

Given that the underlying luminosity curve of a photo is within bounds, correcting a red channel that has hit the wall is best dealt with in an editor/converter which has good colour controls. Then you can select the red area and adjust its saturation and/or contrast as needed.

Underexposing a photo to protect the red channel can introduce noise in all areas, not just shadowed areas - especially when using a camera that has smaller dynamic range.
Andrea G. Blum
Often found hanging out with flowers & bees.

#6 msubees

    Zach Huang

  • Members
  • 389 posts
  • Location: Michigan, USA

Posted 12 June 2014 - 13:41

Andrea and Igor,

sorry, my bad!

1). the correct thickness was 1 mm for UG11 and 1.75 for S8612. There should be no IR leak!
2). I fiddled with the WB and could not really get the same feel I get with D70....but the original might be actually better.
3). The vis one I actually increased the exposure to represent what my eyes saw in the red. so whites are bit over, the original is ok.
4). I reshot anyway because another variety is also blooming (more like an epi).
5). while shooting UV it got me thinking: how to represent the actual darkness/whiteness of a very dark UV flower? without a UV grey scale it will be difficult! so I shot my teflon pad today and hoping to use that as a standard. I tried 1.5" (too dark even for the pad), 3", 6", 12"...etc. I think 6" might be ok. Still both types of red cactus flowers are UV grey, instead of UV black of the queen of the night flowers (Epiphyllum oxypetalum), which I did many years ago with D70.

with visual shots, we can fiddle with exposure compensation. for example if a flower is quite dark, to represent what we saw, we can dial to -2EV. but for UV since we cannot see, this can be a problem! a grey scale will be nice but without it, perhaps we can have a simple black/white card under UV? I guess metering on leaves for some plants might be ok if we know the leaves are UV grey/neutral. but many plants also have UV black leaves..

The problem is, even with visible light shots, we tend to make the photos "comfortable" (e.g. to see details), rather than "actual", so dark flowers tends to be overexposed..we may have to do the same for some UV dark flowers? but how do we know how much is tweaked, and how much is actual?

Edited by msubees, 12 June 2014 - 16:47.


#7 msubees

    Zach Huang

  • Members
  • 389 posts
  • Location: Michigan, USA

Posted 12 June 2014 - 14:06

I resubmitted what my camera saw, no edits at all in photoshop. I converted using Silkpix into Tiff and added caption. No changes in WB, level, saturation, sharpness, etc. The photos were white balanced with a sheet of white teflon, which show slight differences to the virgin one I recently obtained. I shot the virgin one this morning, so I should be able to group select these, and then tell a software what was white, adjusting to the virgin one. i need to finish submitting a real paper today :lol:
it has been sitting there for 3 weeks.

The Epiphyllum I shot many was ago is here: http://ww2.beetograp...-of-the-night-s

it is a quite bit darker. Interesting thing is i shot this at night indoor! so the halogen light I used must have some UV.

#8 Andrea B.

    Desert Dancer

  • Owner-Administrator
  • 5,066 posts
  • Location: USA

Posted 12 June 2014 - 14:26

Zach, good corrections to the shoot.
Good to hear that you are using the 1.75mm thick S8612. That should keep IR-leakage subdued.
And so we have a non-UV-dark cactus flower. Cool !!

...how to represent actually darkness/whiteness of a very dark UV flower?
Well, we do not faithfully represent the actual tonal range of our UV flowers. This is because when converting a raw photo either in-camera or in an editor/converter a (mostly) midtone "gamma" curve is applied to lift the midtones so that the photograph looks normal to human eyes which have a strong midtone lift themselves. The camera records data linearly, but our eyes see non-linearly. If the photo's raw data is extracted without that gamma curve, then the photograph would be very dark - in some cases almost unseeable. You can test what this might look like in your own visible or UV photos by using the dcraw app which has an option for converting a raw photo without the gamma curve tweak.
Unless doing real scientific research which might call for a judgement of the actual tonalities in a UV photo, we simply present our work here as good photographs which have been developed for human eyes. So aim for that in the typical botanical post. The UV tonalities will remain relative even while showing brighter in the midtones than they actually are.

It might be a good idea for me to write up a small post illustrating this. I worked on one some time back but never quite got around to posting it. I'm currently shooting architecture (...me?? shooting architecture?? YAK!!!! I know absolutely nothing about shooting architecture.....) in Boston and Cambridge, Mass. So perhaps I can post this in a few days when I'm back home.

P.S. Don't forget to add your botanical author to the Disocactus post. Read here about Botanical Nomenclature and Author CItations: http://www.ultraviol...l-nomenclature/
Andrea G. Blum
Often found hanging out with flowers & bees.

#9 msubees

    Zach Huang

  • Members
  • 389 posts
  • Location: Michigan, USA

Posted 12 June 2014 - 14:50

Andrea,

Thank you for all the patience and corrections..I added the author for the name. .paper accepted? :lol:

I have another red cactus blooming, it is either a hybrid of Disocactus x Epiphyllum or simply an Epiphyllum. Not sure which. Flower is larger and more open, similar to a white Epiphyllum, and blooms for a shorter time. The one I presented above will last at least one week.

#10 Andrea B.

    Desert Dancer

  • Owner-Administrator
  • 5,066 posts
  • Location: USA

Posted 12 June 2014 - 22:27

The 2nd UV shot is between 1-2 stops underexposed. So it's not clear really where the white balance is going to fall. When I white balanced on the entire photo and added about 1.5 stops on the exposure slider, I got this.
Attached Image: post-41-0-21165600-1402581577-1ProofResize.jpg

The visible red is about 1 stop under. It brightens up nicely in an editor which permits control of highlights. Alternately, mask the white before brightening so it doesn't blow out.
Attached Image: post-41-0-79766700-1402581576ProofResize.jpg

However, please come back to this after you have finished submitting your paper !! :lol:
Andrea G. Blum
Often found hanging out with flowers & bees.

#11 igoriginal

    Iggy

  • Members(+)
  • 582 posts
  • Location: Jackson, Mississippi

Posted 13 June 2014 - 00:13

@ Zach,

Nice correction for the VIS shot! Much better, in my opinion.

As for the re-submitted UV shot, I fear it actually looks even worse, because now it looks monotone. Yellow-green across the entire frame. Just doesn't seem right at all. I think your previous UV-balancing attempt was actually better.

I actually like what Andrea did with your UV photo (trying to WB correct it). It looks very close to what I would expect from a UV neutral-balanced image, because I can now see some grays in the background.

Edited by igoriginal, 13 June 2014 - 00:16.

Igor Butorsky

#12 msubees

    Zach Huang

  • Members
  • 389 posts
  • Location: Michigan, USA

Posted 13 June 2014 - 01:45

Andrea,

thank you! the UV photo now looks really nice. I am not sure how you did it since you do not have a raw file. I thought it was impossible to WB a jpg file. So you must have used channels? I tried many times today and could not get anything close to this. Silkpix does not have a WB selector, it seems. PhotoNinja does but does not apply one selector to many photos (lightroom does that).

#13 msubees

    Zach Huang

  • Members
  • 389 posts
  • Location: Michigan, USA

Posted 13 June 2014 - 14:42

Also, monitor makes such a big difference! My sony laptop at home shows quite yellow of the UV shot. so last night I thought Andrea edited it! my office monitor here (a Dell 30" supposedly calibrated) might be adjusted too dim then. I failed to see the hint of yellow here yesterday but my cell phone also shows quite yellow.

I am adjusting my brightness and contrast to 50 now. it used to be 42. My eyes were hurting at one time I toned them down...

Edited by msubees, 13 June 2014 - 14:51.


#14 msubees

    Zach Huang

  • Members
  • 389 posts
  • Location: Michigan, USA

Posted 13 June 2014 - 15:29

I saved WB in PhotoNinja, then "saved settings to XMP", hoping Photoshop will recognizing it. but it wont open DNG with the WB changes made in PhotoNinja....PhotoNinja does not have an option of opening an image (being edited in PN) in Photoshop, exporting all the parameters to there. Maybe there is another thread I can ask more questions.

#15 msubees

    Zach Huang

  • Members
  • 389 posts
  • Location: Michigan, USA

Posted 13 June 2014 - 16:32

"almost correct" WBed UV shot updated. Used PhotoNinja to open, increased exposure, then picked a place for WB.

#16 igoriginal

    Iggy

  • Members(+)
  • 582 posts
  • Location: Jackson, Mississippi

Posted 13 June 2014 - 16:34

msubees said:

I thought it was impossible to WB a jpg file.

Actually, you can apply white-balancing (using the click-to-white dropper tool) to virtually ANY image file: JPEG, TIFF, PNG, BMP, even GIF.

The issue is how well the resulting quality of the image will be retained by the additional manipulation (Ex: Added noise? Loss of information / details?).

But, just be aware that white-balancing tools can be applied to virtually all / any images. It's just that RAW preserves so much more information, hence more to work with.

So, it's not a matter of if a white-balancing tool can be applied to any given image. The question is, are you starting with the optimal image available to you, for the editing?

(NOTE: The only thing you cannot do with other images besides RAW, as far as white-balancing correction goes, is applying pre-set white-balancing schemes associated with your own camera's white-balance information which is stored within the embedded data of the RAW image. This is because saving to many other images can strip the additional data embedded in the original RAW file which applies specifically to your own camera's hardware and software algorithms. However, a manually-operated click / dropper WB tool can be applied to just about any image which a photo editor can open.)

Edited by igoriginal, 13 June 2014 - 16:45.

Igor Butorsky

#17 igoriginal

    Iggy

  • Members(+)
  • 582 posts
  • Location: Jackson, Mississippi

Posted 13 June 2014 - 16:47

msubees' said:

"almost correct" WBed UV shot updated. Used PhotoNinja to open, increased exposure, then picked a place for WB.

Yep! You just about nearly nailed it, this time around! The updated UV image looks like it's pretty close to a UV-neutral image. Or what I would expect, from my own experiences, anyway (even though I am still largely new to the game, myself).

(Just a bit low in contrast for my taste, though. I would increase contrast just a tiny bit. But that's all up to individual taste, of course).

I would also consider trying to do a highlight-curve reduction on that one blown-out bright spot on the petal to the right (but now, I am just nitpicking, admittedly. Haha.)

Otherwise, looks like you are on the right track! The presence of UV-neutral grays / silvers in the background are the tell-tale sign.

Edited by igoriginal, 13 June 2014 - 16:59.

Igor Butorsky

#18 msubees

    Zach Huang

  • Members
  • 389 posts
  • Location: Michigan, USA

Posted 13 June 2014 - 17:15

Igor,

thank you! Now I notice the blownout spot also for the UV shot. I do not use curve that much and probably do not know how to do it quickly. S shape?

#19 Andrea B.

    Desert Dancer

  • Owner-Administrator
  • 5,066 posts
  • Location: USA

Posted 13 June 2014 - 17:35

I put the Zach's UV foto in Photo Ninja and then dragged the Color Correction dropper across it. Out popped the desired yellow and blue. I then added 1.5 stops on the Exposure slider, dialed the Highlights slider back to deal with the hot area on the right petal. Then I added a bit on the Illuminations slider to pull up the midtones.

Not all editors permit white balancing on a Jpeg. Capture NX2 will not WB a Jpeg, for example. However Photo Ninja is remarkably good at white-balancing a Jpeg. PN is worth every penny!!!

NOTE:
As Dave mentioned somewhere, save the output from working on your raw in Photo Ninja as a Tiff if you want CS3 to see the edits.
Andrea G. Blum
Often found hanging out with flowers & bees.

#20 msubees

    Zach Huang

  • Members
  • 389 posts
  • Location: Michigan, USA

Posted 13 June 2014 - 17:46

Andrea,
Thank you! I did not know you can drag over an area...I used to only choose a spot. Will soon fix the overexposed area also.