Tuesday, April 21, 2009

Bidens ferulifolia - Simulated Bee Vision

Well, and how might a bee now see that flower? Let me show you my "bee vision version" of that flower, compressing the UV information into the human spectral range, i.e. I use the visible and the UV image (cf. my previous BLOG entry on Bidens) and overlay them in a certain way, so as to get the UV-B-G information in one image. As you may see, this is quite a cumbersome process, since ideally one would need two identical shots, one UV and one VIS, to get an exactly matching result, which due to wind, mechanical movement when changing filters etc. is pretty impossible:

[click on image to see a larger one]

Now I was curious to see, if I could have that very same result with just one shot using a suitable filter, and indeed this seems to be possible! I call that filter now my "bee vision" filter XBV:

It is pretty obvious that the "one shot" bee vision image reveals plenty more information, if you look at the petals, the pollen on the anthers etc. No wonder actually, since the overlaying process destroys fine details since a perfect overlay of two shots minutes apart can never perfectly match! And forget that when there is the slightest brise of wind...

For the more scientifically interested (esp. why these tips are so bright, which cannot be seen in visible light with our human eyes), here a reflection measurement I just made (100% = white spectralon) which clearly shows the high reflection of the petal tips in the 300-400nm region, max. around 360nm - the maximum of the bee UV cone sensitivity. Coincidence? I don't think so....:

Stay tuned, more will follow on that fascinating subject...

More info on this very interesting field may be found on my site http://www.pbase.com/kds315/uv_photos