Campanula carpatica 'Rapido Blue' [Carpathian Harebell]

[Hornblende]

Viala, M. (2017) Campanula carpatica Jacq. 'Rapido blue' (Campanulaceae) Carpathian Harebell.

Flowers photographed in visible and ultraviolet light + simulated bee vision. http://www.ultravioletphotography.com/content/index.php/topic/2306-campanula-carpatica-rapido-blue-carpathian-harebell/

 

Ottawa, Ontario, Canada

3 July 2017

Cultivar from grocery store

 

Other Common Names:

• Tussock bellflower

 

Comment:

In the ultraviolet photo, the petals are UV reflective while the stigma and anther strongly absorb UV light.

 

Equipement [Canon 6D full spectrum + EL-Nikkor 80mm f/5.6]

 

Visible Light: [f/16 for 1/320" @ISO100 in sunlight with Lifepixel UVIR-block filter - sunlight white balance preset]

 

Ultraviolet Light: [f/5.6 for 1/50" @ISO6400 - in sunlight with Baader UV-pass filter - white balance on white PTFE]

 

Simulated bee vision (UV-B-G):

 

Close-up in ultraviolet Light: [f/5.6 for 1/60" @ISO6400 - in sunlight with Baader UV-pass filter - white balance on white PTFE]

 

 

[Published 3 July 2017]

[Guest]

Just curious about how UV-B-G is made - and why there is red in that image?

[Hornblende]

Hi Mark,

It is really easy to do with photoshop.

First I take a picture of the flower in UV light and visible light. Then, on the visible image, I assign the green channel to the red channel (the red channel is dismissed), the blue channel to the green channel, and the UV picture to the blue channel.

I then get a UV-B-G image of the flower (instead of B-G-R), which is a simulation of what we could see if, like bees, we had UV-cone receptors, blue-cone receptors and green-cone receptors.

Since the green foliage is assigned to the red channel, the leaves appears red on the pitcure.

 

I don't think bees perceive the same color as on the picture but they can clearly see the flower standing out from the leaves.

[Andrea B.]

We can somewhat predict how the flower will appear to the bee.

For example, the visible purple colour of the Campanula is some combination of Red + Blue (with maybe a small bit of green to pull down the saturation?). The bee has no red receptors, so will only detect the blue reflections from the Campanula. The bee will also detect some reflected UV from this flower as it seems rather UV-reflective except for the central reproductive parts and the veining on the corolla. The combination of reflected UV and reflected Blue is a colour seen by the bee as UVBlue, for lack of a better name.

 

Now, how we reproduce this "invisible" UVBlue in our human trichromatic vision -- well that is a whole 'nother story, as they say. There are a few different models, one of which is the approach you used.

 

There is of course a huge oversimplification in what I've just written. Insect vision is dauntingly complex. But so very fascinating.

 

And it's fun, isn't it? I've always really enjoyed seeing how everyone approaches insect vision modeling.

[Guest]

Okay, thanks. I understand now. Its a false-color representation of what an insect may see, not a reproduction (which is physically impossible for us to 'see'). Sort of like thermal IR images - there's no way we could see that, and the images we generate are just versions of it that are remapped into our visible range. I think that is part of why I like all this UV and IR photography - it lends itself strongly to art (albeit within the constraints of what can measure), since its something we could never actually see, we have some artistic freedom to interpret it into the domain of our sensibility.