Jump to content
UltravioletPhotography

Special Bee Vision Filter


msubees

Recommended Posts

I was chatting on facebook with a friend from Taiwan...telling him that I am into UV stuff. He sent me a link of his own paper saying that they even have filters (set of 3) that matches quite nicely the spectral sensitivity of honey bees! I got all his papers last time we met, but never read them... B) :D

 

the paper is

 

Chiao C-C, Wu W-Y, Chen S-H, Yang E-C. (2009) Visualization of the spatial and spectral signals of orb-weaving spiders, Nephila pilipes, through the eyes of a honeybee. J Exp Biol 212:2269–2278.

 

anyway here is a figure from the paper:

 

post-41-0-72162200-1407512930.jpg

 

Method of fabrication:

 

post-41-0-87173800-1407512941.jpg

 

I will ask him if i get one of these, i will meet him in Taiwan early October.

Link to comment
Do remember you need to take the spectral characteristics and sensitivity of the camera sensor into account as well. Thus, even though the filter may be 'perfect', the outcome will not be unless wavelengths into the visible spectrum are strongly attenuated first (to match the typical lower response in UV even for a modified camera).
Link to comment

Yes, I realize that. In addition, one needs to consider the lens. if the lens transmits less UV (compared to visible light), then their two non UV filters have to be adjusted with neutral density perhaps (assuming it does not affect UV).

 

they used a camera called XCD-SX910UV, SONY Corporation, Tokyo, Japan

Link to comment
I think they took 3 photos of each subject (blue, green, UV), and then post processed to get a photo. So this way they do not need to consider the higher transmission from visible light?
Link to comment
Zach: Of course they would need to take that into account. Otherwise you'll end up with an extremely noisy and poor-quality image. This because UV is the "weakest" link in the chain as it were.
Link to comment

I have read this paper, it is free on the publishers web page. Be sure to also download the 3 supplemental figures not part of the main paper.

While not new is is IMHO a pretty good paper which goes to some extent to "do the math" (literally) for many variables.

Good reading for those of you into insect vision. The images are rather well presented.

I actually have a copy of the 25mm quartz lens similar to they used on their interesting filter wheel UV monochrome CCD camera (supplemental figure 1).

My main issue is with the spectrum of light source used (supplemental figure 2). I think they should have employed a properly filtered large Xe-arc solar simulator.

 

Edit: They used a Pentax H2520-UVM, a now discontinued quartz/fluorite lens

Link to comment

John,

 

good point that they did not use a light source similar to sun light. how does theirs different from that of a sun light?

Link to comment

Zach,

 

Looking at their supplementary figure 2 there are several significant differences. The presence of several prominent Hg lines, the three distinct humps from combining three different lamps one of which emits UV below the solar ozone cut off. The balance of intensity is therefore unlike natural sunlight which is of course the environmentally relevant spectrum for insect visual research.

 

Given that they so diligently controlled so many other aspects of the experiment, I find it a bit surprising that they did not construct a more accurately modeled illumination spectrum.

 

One of my clients manufactures Xe-arc solar simulators, see fig 3 & 6 on this linked page, if you wish to see good spectral match to sunlight.

Link to comment

I have been enjoying this paper. Thanks, Zach, for posting the link.

 

The bee filter idea has always been of great interest. I have had reservations about some aspects of "translating" bee vision to human vision. Don't get me wrong, I certainly encourage and applaud any effort that brings we UV-less human creatures some feeling for how bees see. It is just that bee vision gets first "translated" to camera vision (by necessity, of course) and subsequently stuffed into the RGB model in ways that always leave me wishing for a better model. I should post an example to illustrate a couple of the problems I have with bee vision models. Maybe if I get some time later today I will try to do that.

Link to comment

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...