• Ultraviolet Photography
  •  

Bananas in UV


9 replies to this topic

#1 enricosavazzi

    Member

  • Members
  • 514 posts
  • Location: Borgholm, Sweden

Posted 09 August 2014 - 12:08

Ripening bananas have been reported to fluoresce bright blue when irradiated with UV-A around 355 nm (Moser et al. 2008: Blue Luminescence of Ripening Bananas. Angew Chem Int Ed Engl.; 47(46): 8954–8957. doi:10.1002/anie.200803189; note: you may need to be a researchgate.net member to access this paper). This tells us little about the behavior of bananas at different ripening stages when imaged in the UV.

I decided to test whether there is any noticeable difference in UV images, related to different ripening stages. I did not have available any bananas so green as to be unedible, but I did happen to have a slightly unripe one and a slightly overripe one. Coastalopt 60 mm, Bowens 1500Pro electronic flash with uncoated tube and dome, converted Panasonic G3. I converted all UV and IR images to BW to make it easier to compare them.

Attached Image: P1020338s.JPG
Visible light only, Baader UVIR cut

Attached Image: P1020346s.jpg
Astrodon UVenus

Attached Image: P1020340s.jpg
Baader U

Attached Image: P1020343s.jpg
Asahi Spectra XRR0340

Attached Image: P1020345s.jpg
Omega 325BP10

Attached Image: P1020347s.jpg
850-1,100 nm

Apparently, UV photography is not useful in deciding whether a banana is ripe. Our services are not needed at our local fruit store or supermarket. However, a couple of interesting observations can be made.

The first is that the brown spots are less and less discernible at decreasing wavelengths because UV radiation penetrates less and less through the surface, and is instead reflected by the surface itself. A similar result is observed above 750 nm, but for a different reason: all tissues, including the brown spots, become highly transparent to radiation.

The second is that now we have an alternative to NASA, if we should need a picture of a cratered planetary surface. :o

Attached Image: P1020343c.JPG
1:1 crop of picture taken with Asahi Spectra XRR0340
-- Enrico Savazzi

#2 Andrea B.

    Desert Dancer

  • Owner-Administrator
  • 7,397 posts
  • Location: USA

Posted 10 August 2014 - 13:44

Nice experiment Enrico !! Thanks.
I'm on-the-road so I will get back with more comments in a couple of days.
Andrea G. Blum
Often found hanging out with flowers & bees.

#3 Andrea B.

    Desert Dancer

  • Owner-Administrator
  • 7,397 posts
  • Location: USA

Posted 13 August 2014 - 15:12

I think this cool experiment belongs in UV Forensics where it might be useful to someone someday who is investigating fruit ripeness or whatever. So I have moved it here.

...the brown spots are less and less discernible at decreasing wavelengths because UV radiation penetrates less and less through the surface, and is instead reflected by the surface itself.

Very interesting observation.
Question: I am wondering if the brown spots in the Omega photo are less visible because those brown areas become reflective in shorter UV rather than because of UV penetration depth ?? (I hope I am phrasing this question meaningfully.) I ask this because it appears that the indentations are still the same in the Omega UV photograph, so we still have a sense of seeing the same depth as in the Baader-U photograph.
Andrea G. Blum
Often found hanging out with flowers & bees.

#4 enricosavazzi

    Member

  • Members
  • 514 posts
  • Location: Borgholm, Sweden

Posted 14 August 2014 - 09:33

Reflectivity at the surface is a different property than penetration through the structure below the surface, which depends on absorption, reflection, scattering etc. within the structure. In this sense, the question certainly does make sense. It is much more difficult to give a meaningful answer, especially without access to sophisticated equipment.

What the pictures seem to display is that the brown spots become progressively less visible with decreasing wavelengths. The three-dimensional relief of the surface of course remains he same, but becomes more obvious as the differences in absorption fade out. This is the same effect that we obtain in scientific photography when we coat specimens with ammonium chloride or magnesium oxide before imaging to eliminate color and translucency, and leave only the surface topography. This is also similar to what happens when we gold-sputter SEM specimens. Incidentally, this is also the cause of the "golden reflections" of particularly smooth plant surfaces imaged around 350-360 nm (together with the fact that Bayer sensors record these wavelengths as "yellow").

Based on my qualitative interpretation of the pictures, it seems that reflectivity at the surface increases (the size and intensity of the specular highlights seems to increase significantly), at least relative to internal absorption by brown spots. This increase in reflectivity can effectively mask anything else that is going on under the surface, so I would not be able to say whether the dark (=absorbing) quality of the spots actually decreases at decreasing wavelengths, or is simply hidden under the more reflecting surface.

Edited by enricosavazzi, 14 August 2014 - 09:37.

-- Enrico Savazzi

#5 Andrea B.

    Desert Dancer

  • Owner-Administrator
  • 7,397 posts
  • Location: USA

Posted 14 August 2014 - 18:01

.....so I would not be able to say whether the dark (=absorbing) quality of the spots actually decreases at decreasing wavelengths,
or is simply hidden under the more reflecting surface.

We have posted many examples of flower petals which appear UV-absorbing when the UV illumination is directed in one angle but which also have UV-reflective areas when the UV illumination is aimed in another angle across the surface of the petal. (For readers not knowing - this dark/light duality occurs because of conical cells on the top of the petal which reflect or scatter light.) Usually it can be decided whether the petal actually is UV absorbing by playing a UV torch at different angles across the petal and viewing the changes in Live View or a short video sequence.

So would that work with your ripe banana spots ? Using different illumination angles to determine whether the spots' UV reflectivity is dependent upon that ? My initial guess is that the answer is no and that the UV absorption curve of the spots has a jump in the 350-360nm region.
But it is just a guess. :D

**********

....the cause of the "golden reflections" of particularly smooth plant surfaces imaged around 350-360 nm (together with the fact that Bayer sensors record these wavelengths as "yellow").

Perhaps if a certain white balance is set in certain cameras, this would be true. :)
What white balance are you indirectly referencing with this statement? Using the Panasonic G3, you have probably set an in-camera white balance to see your yellows in the 350-360nm range, I'm guessing? What standard did you use to set the white balance to get these yellows ?

False colours are sooooooo tricky to reference because they are dependent on so many variables - the camera sensor, the camera software, the Bayer filter, the lens, the filter, the illumination, the camera white balance both native and adjusted, the editor used to convert the digital photo. Sometimes I wish we simply presented all our UV photos in Black & White only as you did above. :D Of course there could be variations in B&W too so it may not matter. I'm amazed that we are able to get a certain almost similar "look" in our UV photography across all these variables.
Andrea G. Blum
Often found hanging out with flowers & bees.

#6 enricosavazzi

    Member

  • Members
  • 514 posts
  • Location: Borgholm, Sweden

Posted 15 August 2014 - 10:16

View PostAndrea B., on 14 August 2014 - 18:01, said:

Perhaps if a certain white balance is set in certain cameras, this would be true. :D
What white balance are you indirectly referencing with this statement? Using the Panasonic G3, you have probably set an in-camera white balance to see your yellows in the 350-360nm range, I'm guessing? What standard did you use to set the white balance to get these yellows ?

False colours are sooooooo tricky to reference because they are dependent on so many variables - the camera sensor, the camera software, the Bayer filter, the lens, the filter, the illumination, the camera white balance both native and adjusted, the editor used to convert the digital photo. Sometimes I wish we simply presented all our UV photos in Black & White only as you did above. :D Of course there could be variations in B&W too so it may not matter. I'm amazed that we are able to get a certain almost similar "look" in our UV photography across all these variables.
In general, I use a custom in-camera WB that gives approximately correct VIS results with a converted camera and a UV and IR cut filter, like the B+W 486 or Baader UVIR Cut, and a given light source (usually electronic flash). These two filters are not exactly the same and require a slightly different WB. Then I go on shooting with the same WB also with the UV-pass filters. In other words, I use the same in-camera WB for all UV, VIS and NIR images, because typically I shoot several UV images of the same subject with different filters and it would be too time-consuming to change in-camera WB for each filter. I am aware that others, including on this BB, do things differently.

I use a "white" PTFE target only when I want to have a fixed reference, which I find more practical for WB in PP. I do additional WB in PP only when using a (presumably variable) solar/ambient illumination and when the results need to be comparable among different images.

To convert UV images to BW, I don't do it the "Photoshop way". Instead, i simply turn color saturation to zero without applying a differential weight to each color channel.

Bayer sensors seem to respond in quite similar ways in terms of UV false color, regardless of make and technology. I suspect that the dies used in the Bayer color filters are pretty much the same throughout the industry. All bets are off with radically different sensor types, like the Foveon. Sascha Hein is doing interesting quantitative work on the spectral response of two different models of converted cameras with Bayer sensors, and eventually we are going to release the information, but it is not ready yet.

Edited by enricosavazzi, 15 August 2014 - 10:17.

-- Enrico Savazzi

#7 Andrea B.

    Desert Dancer

  • Owner-Administrator
  • 7,397 posts
  • Location: USA

Posted 15 August 2014 - 15:52

It is always very informative to hear how another UV photographer handles white balance and other aspects of UV photography. So thank you for this informative input, Enrico.
***

To convert UV images to BW, I don't do it the "Photoshop way". Instead, i simply turn color saturation to zero without applying a differential weight to each color channel.

Yes, I do this also. We have to be careful not to alter the relationship between the tones in UV documentary work, I think. Although I will sometimes adjust the B/W points - or sometimes open up the shadows a little bit to show detail - in either B&W or color UV photos.
***

Bayer sensors seem to respond in quite similar ways in terms of UV false color, regardless of make and technology.

It does seem to be so, thus far. Still, we must be careful of imputing too much meaning to false colour I think.
***

Sascha Hein is doing interesting quantitative work on the spectral response of two different models of converted cameras with Bayer sensors, and eventually we are going to release the information, but it is not ready yet.

That's great!! We really need this kind of information.
***

...for WB in PP
Sorry, I don't know editor this is?
Andrea G. Blum
Often found hanging out with flowers & bees.

#8 enricosavazzi

    Member

  • Members
  • 514 posts
  • Location: Borgholm, Sweden

Posted 16 August 2014 - 07:28

(I meant PostProcessing) :)
-- Enrico Savazzi

#9 Alaun

    Member

  • Members(+)
  • 284 posts

Posted 16 August 2014 - 13:39

Going more into details :) reveals some more interesting substructures ...

(very ripe one, not like in the background, Pany GH3UVIR, Noflexar 35, BaaderU, SB14mod)

Attached Image: _1020637_v1_1200.jpg



Werner
Werner

#10 Andrea B.

    Desert Dancer

  • Owner-Administrator
  • 7,397 posts
  • Location: USA

Posted 16 August 2014 - 14:31

Werner, cool shot. Those fibers must be why bananas are supposed to be good for us aside from the vitamin content.
Now I gotta go out and buy some bananas.
Let's see some work with apples, peaches, strawberries and so forth! "-)

aha PP = post processing = editing. Thx. I thought there might be a new editor I didn't know about. "-)
Andrea G. Blum
Often found hanging out with flowers & bees.