Jump to content
UltravioletPhotography

Misumena vatia [Crab Spider on Buddleia]


Recommended Posts

I have wanted for a long time to photograph a crab spider in UV, and finally got the chance today, with the specimen which has been living on the Buddleia on my patio. Interesting how the contrast completely disappears in UV - no surprise it catches lots of flies!

 

Nikon D70, with 50mm El Nikkor, Baader U filter. 1/2 second f/5. for the UV exposure in daylight.

post-47-0-11522100-1409674302.jpg

Link to comment

Cool that you were able to photograph the Spidey!! These little guys fascinate me and I love to photograph them too.

 

But may I suggest that before making any judgements about its appearance in UV, it would be a good idea to reduce the red over-saturation in the UV photograph which obscures both detail and tone. If you don't want to white-balance the photo, then use the highlight slider and contrast slider in some editor to pull the red spike off the right-hand histogram wall.

Link to comment

Thanks for your helpful comments. I obviously need to start looking n

more closely at colour issues. Where dos the red colour come from? Would a "hot mirror" filter help (to absorb infrared)?

Adrian

Link to comment

The red colour is mainly due to the response in UV. Do remember that we are using these cameras *far* outside their design envelope, so results are less intuitive or predictable. Thus, red channel responds to UV, but also IR if you have insufficient protection against IR.

 

There are several articles on UVP on how to properly w/b the camera to give a more predictable and repeatable false-colour palette. This w/b can be achieved in-camera with some models, others may need the setting to be conducted in the RAW conversion stage. Ensure you are safe-guarded from IR contamination if you wish to record only UV features, by using the appropriate filter(s).

Link to comment
As Adrian was using the Baader-U, the photo is definitely UV. If you open the D70 raw photo in VIew NX or Photo Ninja or Capture NX2, find the white balance dropper and then click a magenta area of the photo, then it usually will pop into a useful white balance which can be further refined if desired.
Link to comment

Adrian, here is a quick edit just to show you approximately how the photo would look after white balancing.

post-47-0-11522100-1409674302pn.jpg

 

And in this quick edit black and white version, you can see that your original conjecture is correct - the crab spider does not stand out strongly against the buddleia flowers in UV like it does in visible light. I wonder if this protects the spider from being eaten by birds??

post-47-0-11522100-1409674302pnpnCropResBW01.jpg

Link to comment

...the crab spider does not stand out strongly against the buddleia flowers in UV like it does in visible light. I wonder if this protects the spider from being eaten by birds??

 

Or perhaps helps camouflage the spider from prey insects?

 

Might be cool to see what spidey looks like in UV/B/G insectovision.......

Link to comment

It would be cyan against mostly blue flowers.

Visible white which is UV absorbing = +R+G+B-UV --> +G+B-UV = bee blue-green.

Link to comment
I wish predicting UV false colours would be that simple :D. From what examples I've seen posted elsewhere over the Internet, colour consistency is not what first comes to mind. I've just ordered one of these "bug" filter stacks to find out more on my own.
Link to comment
There are many types of camouflage. Camouflaging as bird excrements (usually white with dark spots in the VIS) is not unheard of. It would be interesting to know if these spiders match the UV reflectivity of bird droppings.
Link to comment

I wish predicting UV false colours would be that simple

 

Not a "prediction" really. That analysis was taken from a whole lot of papers about bee vision.

q.v.

 

I have no idea how the so-called insect vision filters work. Such filters would have to take into account the proportion of green and blue relative to UV in sunlight if they are to fairly represent bee vision. At this point I am still thinking that making channel stacks in Photoshop gives the best representation of bee vision. But that of course requires the use of the Red channel which makes the final photo slightly weird.

 

There are of course nuances in simulating bee vision. We can easily simulate UV-absorbing colours in bee vision. They are simply green, blue and cyan. With possibly some yellow nuances depending on how far the green shoulder extends. But it is the UV reflecting colours which are difficult to simulate. My current model for this is to use dark-green, dark-blue and dark-cyan for the UV-absorbing areas. And light-green, light-blue and light-cyan for the UV-reflecting areas. It requres of course some hand painting in Photoshop which puts most people off. :D

 

********************

 

Good question Enrico. We occasionally get some owl "whitewash" in my yard when the Great Horned Owl stops in the yard to hunt for rodents. I will put it on the list to photograph in UV.

Link to comment

A filter combination that tries to balance a capture in UV, blue, and green. As the camera's response is very uneven between these bands, in particular UV compared to the visible bands, the transmission curve is strongly biased towards the UV band and attenuation of the visible bands is quite massive. Thus, the filter factor is huge as well.

 

A potential issue the bug filter users never seem to discuss is what kind of colour balance applies to such captures. I hope to address this later when my filter arrives.

Link to comment

Try eBay seller 'uviroptics' - he has several versions either as combined filters or combinations of UV-pass (UG-5 which transmits also in the visible) and S8612 (which blocks IR and even deep red depending on its thickness).

 

While my insights into these filters are purely theoretical at present based on published transmission curves), my hunch is they will work better for a lens that transmits freely in the UV range. As soon as my filters arrive I'll test this assumption of course.

Link to comment

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...