Using Hoya U340 for landscape photography

[DonPilou]

Hello everyone!

 

When I started reading about UV photography, I quickly bought a Hoya U340 filter with the biggest diameter I found (82mm from UVIR Optics), thinking that it would be possible to capture ultraviolet with all my recent lenses. Such a mistake... Now that I have spent some money on dedicated lenses and filters, I have tried to use this filter for landscape photography whithout other filter to reduce the IR polution.

 

I think the results are interesting to show them here. To understand how I have obtained this effect, let me explain my work :

• Gear : Canon 6D full-spectrum + Canon 24-105 F/4 IS (UV_1) or Canon 16-35 F/4 IS (UV_2) + Hoya U340
  
• Exif : UV_1 : 28mm, F/10, 5s, 100iso // UV_2 : 16mm, F/14, 0.8s, 100iso
  
• White balance made with a gray chart
  
• Tripod used because of the long exposure
  
• Digital processing : no channel swap, just an important reduction of magenta in the blue using "Color Correction" in Photoshop.
  

As you can see, because of the lenses coat, IR effect is more visible than UV effect : the leaves are bright, the sky is very dark and contrasted... Well there is no UV effect visible. The small part of information given by the left part of light spectrum (around 400nm) allowed the apparition of green colour on the leaves and other parts where the IR effect is applied.

 

So with this filter, you can obtain false color infrared shots with green effect.

[colinbm]

Good show

The first reminds me of the Emerald Isle :D

Col

[Cadmium]

Quite nice photos! I especially like the top one. A lot of people use the UG1 or U-360 as a dual band filter, but it is rare to see someone using U-340 or UG11 for dual band like this, and these are great examples.

Given a lens that transmits UV well enough, and stacking the U-340 with some BG glass will give you a UV-only shot.

Depending on the thickness of the U-340, 1mm or 2mm, then I would use S8612 1.75mm-2mm or 1.5mm-2mm respectively.

You could even try using BG40 2mm with U-340 2mm, but that's pushing it a bit. You want to keep the Red/IR peak down under 1E-03 and preferably about 1E-04 or under.

Again, your shots here are quite splendid. Such a nice use of U-340.

[OlDoinyo]

Very nice first photo, lovely textures and tones. I would not have expected the green foliage, as Woods-glass type filters are quite opaque to actual green light; but somehow the IR is coming across as green. Most of the dual-band examples I have seen with such filters produced straw-colored or reddish foliage.

[DonPilou]

Thanks to all of you for your answers.

[Andy Perrin]

I love these photos! Sometimes it's fun just to play with filters and see what kind of results you get. Strictly UV or IR photos have their place, but from an art standpoint, the final image is what matters.

[igoriginal]

Very nice first photo, lovely textures and tones. I would not have expected the green foliage, as Woods-glass type filters are quite opaque to actual green light; but somehow the IR is coming across as green. Most of the dual-band examples I have seen with such filters produced straw-colored or reddish foliage.

 

If the UV-pass filter glass is thin enough (and, more importantly, there is no IR-blocking glass stacked with it), then of course you'll get a "dual-band" effect with such a glass, especially if you manipulate the white-balance to elicit a particular color scheme, such as the one above.

 

But dual-band images can be manipulated to virtually almost any color scheme between sky and foliage.

 

Yes, you're right that there isn't any significant amount of "green light" getting through (with a typical thickness of UG-340), but that's not what's making the grass appear so vividly green in the case of the above photos; it's the CWB (custom-white-balance) adjustment that's doing it, plus additional post-photo processing.

 

We have to remember that a full-spectrum camera can either record TRUE green (corresponding to the range of 495 – 570 nm), or it can record a FAUX green, depending on custom-white-balance settings and/or post-photo editing. Sometimes, a mixture of both, depending on the optical glass (and thickness) used.

 

Now, the typical "reddish"-tinted foliage that we normally get with a full-spectrum camera is due to the white-balance left in "AUTO" mode, not CWB (custom-white-balance) manipulated.

 

Speaking of which, here is an example of yet another color scheme that a CWB setting can elicit via dual-band use, such as my image below, in which I get a green sky and purple foliage. Quite alien-looking, isn't it?

 

[igoriginal]

Very nice shots, by the way, DonPilou! Excellent work.

[OlDoinyo]

....especially if you manipulate the white-balance to elicit a particular color scheme, such as the one above....

.....But dual-band images can be manipulated to virtually almost any color scheme between sky and foliage.

 

But the white-balance in the photos is constrained by the necessity of having the waterfall or the clouds rendered as white, and the rocks and other objects as grey. You cannot have any arbitrary white balance and have that be true. You could, of course, get greenish foliage via a GRB display intent, but we are not told that this is the case. That is what puzzles me.

[igoriginal]

But the white-balance in the photos is constrained by the necessity of having the waterfall or the clouds rendered as white, and the rocks and other objects as grey. You cannot have any arbitrary white balance and have that be true. You could, of course, get greenish foliage via a GRB display intent, but we are not told that this is the case. That is what puzzles me.

 

The reason that the clouds, rocks, and waterfall remain generally truer to their color, in this case, is because the custom-white-balance itself was set to a color-neutral target (the "grey card" as the target, as was described above). Clouds happen to be color-neutral as well (as potential white-balance targets). This is why there is no strong "off-color" rendered in the clouds, after the CWB was set to a grey card. Ditto for the waterfall and all other color-neutral objects.

 

Using color-neutral white-balancing targets permits for this, even as other objects with vivid and vibrant colors can have their color-schemes swapped at will and displayed in virtually any alternate color scheme, while grey-scale objects remain generally unaffected in their color rendering. There is nothing puzzling about this, if one understands that.

 

Ex: Notice how the rocks in the first scene also remain true to their color, since grey-colored stone is also a color-neutral white-balance target. (In fact, this also remains true when using a color-neutral target for custom-white-balance operation in UV photography - be it a slab of PTFE / teflon, or even a light-grey concrete sidewalk. I happen to use sidewalk concrete as a color-neutral target, when I don't have a dedicated color-neutral target handy).

 

Now, when one uses a non-color-neutral target for custom-white-balancing (such as a vividly red, blue, or green color target), that is a different story, of course. This can cause even grey-scale objects in the scene to acquire a substantial off-white color balance.

[JCDowdy]

Of course the standard gray card may not be a neutral target outside of the visible either.

[igoriginal]

Of course the standard gray card may not be a neutral target outside of the visible either.

 

That's right. Lots of other variables, there. The chemical properties of the surface being used, including various brighteners and other modifying properties. Just because the surface "looks grey" to the human eye doesn't necessarily mean that it's true (neutral) grey. Thus, often, a trial and error is required, with regards to any suspected "neutral" target determination outside of already well-known and established good targets.