Flowers with two false colours

[Cadmium]

Andrea, you are using the 340 on the lens. You have to use it like the Sparticle to see the difference from PTFE white.

It is not about the camera. Every camera I have used dose the same thing.

Get a few 12.5mm bandpass filters from Omega, I already told you would will give you a holder for them.

 

I don't see green or turquoise in reflected things. I don't see yellow in almost anything except a flower. We have not found man made materials that reflect yellow or green UV.

If I use a camera with no lens, light looks more turquoise.

The main idea of the Sparticle was not to see colors, but to measure the UV depth/cutoff point to which a lens transmits, the colors were just a byproduct of that.

The colors are real. I mean as far as the Bayer layer goes. They are 10nm wide colors.

It has absolutely nothing to do with D600 or any other model.

Do this: Take your 340 filter, prop it up somewhere, shoot a pic with it and some PTFE next to it, and some scene/landscape behind all that jazz, and then white balance it, and you be the judge.

The bandwidth of the 340 will have something to do with the color, but you would absolutely get green if the bandwidth were 10nm.

Trust me. :-)

[Andrea B.]

The original idea (from many many many years ago) is that we want to make a false colour UV photograph from which we can determine the reflected wavelength(s). So I must shoot through a filter on the lens.

 

But I will for sure try your suggested experiment.

 

---

 

Another interesting tidbit: Landscape scenes often produce false greens (dark). But close-ups of leaves and grass do not always produce false greens. Scattering effect at a distance?

[Cadmium]

Yes, you shoot through your UV filter (Baader).

This idea is simple.

Go look at Colin's version, I can't find it on here right now.

[Cadmium]

This is just so simple.

[Andrea B.]

Will do!

 

I have the dentist thing tomorrow, so will probably not get back to UVP until next Monday or Tuesday.

[Cadmium]

I prefer the 12.5mm diameter versions, because they fit my holder.

http://www.ebay.com/...40BP10&_sacat=0

 

Get one, taker a pic of it against the sky, with your Baader on your lens, with some PTFE in the pic.

Which balance it, and you will get green.

 

and, I wish you all the best with the dentist.

[Alex H]

Andrea, are you using color-profiled monitor?

[Cadmium]

Here is Colin's Sparticle version page, see the pic at the bottom of his page, it is white balanced, some of the same 10nm wide bandpass filters, 340 is green, etc..

https://www.flickr.c...57644733838706/

 

Here is where he originally posted the link on here:

http://www.ultraviol...__fromsearch__1

 

And I will say this one more time:

DO NOT put the bandpass filter directly on the lens,

INSTEAD you put your UV-only filter/or stack directly on the lens and then take a photo of the bandpass filter(s).

Include some PTFE in the shot as well if you want.

Then white balance the photo just as you would any UV photo.

That is how it is done.

[Andrea B.]

Will do. Promise! :D

 

Just have to take this dental hiatus for a couple of days. Off now to get it over with.

[Alex H]

Just have to take this dental hiatus for a couple of days.

 

Your dentist must be using some strong tranquilizers if they last for so long

[Andrea B.]

Here is the follow-up from Raw Digger. I gave it a separate post for easier future searching.

http://www.ultraviol...-with-response/

 

The summary I wrote there is repeated here:

 

So UV false colour palettes work only on a per camera basis because of the Bayer filters. (In case anyone had not yet observed that.) If you want to try to reverse map false colour to wavelength, you must correlate wavelength-to-response for your individual camera+lens+filter combo in use.

 

A standardized (Bayer filtered) colour palette for UV false colour -- like we use here in the botanical section -- will never be perfect. There will be variations.

(It still amazes me that we get "similar" colours across such a wide variety of gear.)

 

******

 

Alex, lol!

We get no tranks. But I so dislike having the throbbing jaw that I just planned in advance a 3 day holiday from the internet. I hope it won't be as bad as last time.

[JCDowdy]

John, we lack data about sensors' RGB response curves between 300-400nm and ..... one or two measurements without a full description of methodology does not a certainty make.

 

Here is a link to a paper in the SPIE digital library showing a plot of QE of a an RGB CFA that extends down to 350nm. This is the shortest wavelength range I can recall finding.

 

Here is the plot in case that web page is not persistent

 

When I went looking for the methodology, the links to the source for that plot cited in reference 6 & 7 were dead. Google says that company was acquired by another company and that they made the sensors for some of the little Nikon 1 cameras.

 

Added

 

While I am logged in, here are a couple of somewhat relevant YouTube videos someone may enjoy.

 

 

 

Hope your dental isn't to much of an ordeal.

[DaveO]

John,

 

That looks sort of like the D80 spectrum which we got from somewhere:

 

 

I think the problem is not so much the low response but also the fact that the RGB curves cross over each other and ones for different RGB dyes would almost certainly have different crossing points.

 

The false colours are due to the small contributions from the B and G channels otherwise it would be monochromatic. Chemists are good at "arm waving" to describe things so here goes.

 

The false-blue near 400 nm (going down to shorter wavelengths) is due to having response from mainly R and then B. Below about 360-ish where B and G cross over we get response from both B and G, with more from G than B. Guess what happens when you mix R and G: Yellow

 

Dave

[Hornblende]

See the "sensitivity" part from this website : http://www.randombio.com/uv.html

[nfoto]

The author of that reference takes a lot of alleged 'UV myths' to task not realising most of his results and conclusions are due to inadequate technique on his part.

 

"This is another example of how not all flowers have a dark center in the ultraviolet. In fact, the entire flower is dark, yet bees have no problem finding them. Neither do deer."

 

 

" ... many flowers are uniformly dark when photographed in ultraviolet light, which casts doubt on the popular myth that the dark center is a visual cue for bees"

 

" .... daisies are uniformly dark in ultraviolet light"

 

" ...despite the pink or violet color, rhododendron flowers are so dark in the ultraviolet they are almost invisible. Even so, they are always surrounded by bees, casting further doubt on the bees-attracted-to-UV theory."

 

Floral features such as conical cells, iridescence, fluorescence, etc. apparently are unknown concepts? For many years the diversity of UV signatures and the existence of alternatives to the "bull's eye" are well documented.

[Andy Perrin]

He must have written that quite some time ago. When I first started UV photography (around 2013 or so, well before I joined this board) I remember finding the same page in its current form.

[nfoto]

The site says "Update April 2013".

[Andy Perrin]

Sounds about right. :)

[DaveO]

I begin to think that I have been looking at UV through the wrong end of the telescope. Instead of asking why we get UV dark areas in the centres of flowers, perhaps I should have asked, "What makes some flowers REFLECT in UV to give what we see as false-Yellow". I'm sort of coming to the conclusion that the UV-dark centres are there primarily to protect the genetic material in the plant which resides there, there are ample biochemical pathways to UV-absorbing flavonoids as in a recent post of flower colours. So, evolution, over the eons produced plants with UV-dark stamens etc which propagated through the population because they survived better than those without the UV absorbers. Then, insects being infinitely cleverer that we think, learnt to look for the dark areas to get their nectar or pollen rewards.

 

I'll post this in a more considered when I've collected more "evidence" (same day ;) )

[Andy Perrin]

One thing I would like to know is whether the UV yellow we see in the flowers is a physical color (produced by some repeating structure in the petals causing interference effects) or a pigment (purely absorptive). If it's the former, that explains why we see so little yellow in the surroundings outside of flowers.

 

Edit: This was interesting.

[DaveO]

I went looking for some data I had on modern sensors and found one of the links had changed to this:

http://www.onsemi.co...682&lctn=header

 

Which then produced this for a CMOS sensor

 

So we still don't get any data further into the UV because the makers are not interested outside the visible region.

[OlDoinyo]

Someone needs to take converted cameras and put them in front of a tunable laser or monochromated UV source if we want this information. I don't think manufacturers are going to bother researching these out-of-design-envelope performance matters.

[DaveO]

Andrea,

 

In this latest post of mine on Hibbertia crinita

 

http://www.ultravioletphotography.com/content/index.php/topic/2107-hibbertia-crinita/

 

you will find reflectance spectra from Dyer which peak at around 350 nm corresponding to the UV-yellow I got in my photos.

[JCDowdy]

Dave

 

I was scanning through some files I had on sensor spectral sensitivity and found another plot I forgot I had showing data down to 300nm.

 

 

This is for a small c-mount BasleracA2000-50gc

[DaveO]

I bet you would get some weird false-UV colours out of this with the way the curves seem to be crossing each other below 350 and the red seems to be taking more of a dive than in the other curves. Just goes to show that 'Bayer dyes ain't Bayer dyes' to misquote 'oils ain't oils'