Night GBU in town

[OlDoinyo]

A couple of evenings ago, I took the camera out in the gloom and fog to record some street scenes in GBU. The Sony A900 was used with the Steinheil Cassar-S lens at f/16 and ISO 100, capped with either the Kolari deconverter (visible) or the Baader U2 (ultraviolet.) The results follow:

 

 

 

 

Light sources (such as LED) which are visible-white but produce little or no UV show up distinctly yellow, as might be expected. More whitish sources are most likely some form of mercury- or sodium-vapor lamps. The McDonald's arches turn orange-red. There is undoubtedly some IR contamination in the U->B channel, however. Also, it is extremely difficult to white-balance a GBU image at night, due to the lack of any sort of reference target in the images; what you see is inevitably approximate or even arbitrary.

 

I close with a plain UV shot of the Hooterville Pig sign. Display intent is BGR.

 

[Stefano]

I notice that the colors are strange for UV photographs, but as you said there can be IR contaminations (I think that I would have used a filter with at least OD 5 to do this), and you had trouble to WB in almost total darkness. This is what I would have expected:

-Blue-indigo for LEDs (they emit very little UV, and the UV they emit is definitely towards longer wavelengths. Probably nothing below 380 nm.

-Strong yellow for mercury lamps, because of the 365 nm I-line (the 405 nm H-line should be blocked by the Baader U).

-Definitely NO reds.

-Greens are also strange, they should appear in deep UVA and UVB, around 340 nm. I don't think there is much of this wavelength at night, also because mercury doesn't have any strong line there.

[OlDoinyo]

GBU contains visible as well as UV--full mapping is GBU-->RGB.

[Stefano]

OK, so white means equal amounts of UV, green and blue, yellow means equal amounts of green and blue with no UV, green means only blue, red only green and finally blue only UV. Now everything makes a lot more sense. So you have yellow LEDs and red McDonald's sign (which is normally yellow, and this means that it contains green, no blue and no UV). And does the white color of some lamps mean that they emit a significant amount of UV? They should be filtered against UV, for safety and "discoloration" reasons (UV tends to damage plastics and paints, so you don't want it on a plastic cover or something like that).

[Andrea B.]

Do the blue/purple streaks indicate that some of the car headlights are halogen?

[Stefano]

Do the blue/purple streaks indicate that some of the car headlights are halogen?

Halogen lamps should emit good amounts of green, less of blue and even less of UV. They should appear of the same orange shades of incandescent objects. It is like shifting the peak wavelength to longer waves (using a cooler object). A blue objects should be UV-only in this rendition.

[OlDoinyo]

Although the UV image from a digital camera picks up some IR, I know from night film photography (which does not) that whatever filtration these sources (Hg, Na, fluorescent, neon) have is imperfect; otherwisr, photography such as this would not be possible.

 

Halogen and other incandescent sources emit a bit of UV but also a lot of IR; their presence in these images may be due more to the latter than to the former.

[Andrea B.]

Thanks, Clark.

I always enjoy your explorations.

The Hooterville Pig is a fun photograph. Pig appears to be floating through the sky.