• Ultraviolet Photography
  •  

Lots of dandelions

White Balance
24 replies to this topic

#21 Andrea B.

    Desert Dancer

  • Owner-Administrator
  • 8,762 posts
  • Location: UVP Western Division, Santa Fe, New Mexico

Posted 20 March 2021 - 20:56

The dandelion centers are a bit light or bright or whatever the appropriate term is when compared to most dandelion photos I've made. But not by much, so don't read too much into my remark, OK?

In the first photo in Post #16, the centers of the dandelions range from threshold 16 to threshold 64 which is an approximate brightness range between 6% - 25% 6.25%-25%. If we assume brightness can serve as a proxy for UV absorption, then this seems to be "dark enough" for all practical purposes of finding UV signatures. :grin:

I could not find any (0,0,0) pixels in the centers of the dandelions in that photo. I ran a threshold of 8 over the photo. This represents a 3.125% brightness or pixel value of, obviously, (8,8,8). The magenta areas are 0 - 3.125% bright (well, dark in this case). No magenta areas are in the center of the flowers, so no (0,0,0) there.
Attached Image: threshold8.jpg


It is probably totally WRONG to evaluate tonalities on a JPG. But maybe we are in the ball park.
Right now I would like to be in a ball park on a nice spring day.


A Ramble:
But I'm thinking that we don't really know just how dark those dandelion centers are "supposed" to be because we look at our UV photos with a brightness curve applied for human eyes. Although if we set up a photo using a range of UV standards next to one of the flowers, we could make a better determination of the absorption of the dandelion center. Even though the brightness curve would have been applied also to the standards, one could still compare the dandelion center to the standards. That seems like it would be a reasonable approximation of the actual UV absorption measured spectrometrically. However, this "eyeball" comparison of darkness would not have good granularity because some tonality is squished when the curve is applied.

"squished" is a very useful scientific term.
humor alert



Edit: typo in values.

Ah well....Stefano forgive me for rambling on. I don't want to go clean the house which seems to be terribly full of dust from the recent winds here. It is a LOT more fun to play around with thresholds of dandelion centers.
Andrea G. Blum
Often found hanging out with flowers & bees.

#22 Stefano

    Member

  • Members(+)
  • 1,796 posts
  • Location: Italy

Posted 20 March 2021 - 21:41

I didn't actually search for 0, 0, 0 pixels, but it did look to me that the blacks were too black.

#23 Stefano

    Member

  • Members(+)
  • 1,796 posts
  • Location: Italy

Posted 20 March 2021 - 21:43

Andrea, keep in mind that the images were taken with two different cameras. The image you were initially referring to, I think this one, was taken with my old point-and-shoot.

#24 Andrea B.

    Desert Dancer

  • Owner-Administrator
  • 8,762 posts
  • Location: UVP Western Division, Santa Fe, New Mexico

Posted 20 March 2021 - 22:44

No, I was referring to the one I posted above with the Canon and the ZBW stack.
Andrea G. Blum
Often found hanging out with flowers & bees.

#25 Stefano

    Member

  • Members(+)
  • 1,796 posts
  • Location: Italy

Posted 20 March 2021 - 23:03

Well... this is the post you originally quoted: https://www.ultravio...dpost__p__34981

I wrote that in reference to the photo in that post, which of course was taken with my old camera. I don't think my blacks are clipped in the new photos.