Jump to content
UltravioletPhotography

Panasonic Lumix DMC G3 Camera


colinbm

Recommended Posts

Hi John

Both of the above flower shots are diffused CWB on blue sky, with UVIROptics PTFE, over the lens.

Col

Link to comment

I ran that last shot thru Raw Digger for Col.

 

Here is the raw without profiling (until the last step which turned it into a displayable jpeg).

P1070726_RawDigger.jpg

 

Here is the channel breakdown. Under the combo of UG5 + S8612, mostly blue was reflected as you can see. Then Green and lastly Red.

 

Blue Channel

P1070726_RawDiggerBlueChannel.jpg

 

Green Channel (Only one shown. They were both the same.)

P1070726_RawDiggerGreenChannel.jpg

 

Red Channel

P1070726_RawDiggerRedChannel.jpg

 

Here is the Raw Digger Histogram of those channels.

P1070726-RawDiggerFullHisto.jpg

 

The G3 in-camera white balance pretty much matches the Raw Digger auto white balance.

 

G3, as shot.

P1070726_PhotoMechanicAsShot.jpg

 

Raw Digger auto wb.

P1070726_RawDiggerAutoWB.jpg

 

The white balance was taken as far as it could go in-camera. I could get no further in Photo Ninja, Iridient Developer, or SilkyPixw for RW2.

 

When I tried to go "cooler" in Iridient, everything turned blue. No surprise there as blue is the predominating reflected colour in this photo to start with.

 

To white balance this particular photo, I think that perhaps blue would actually have to be pulled back somewhat and red pushed up a bit (under the assumption, of course, that there is actually something neutral in a UV foto to start with. Who knows how the camera is deciding that??? Tis a mystery.) And, as you see, pulling back that blue brings out the green and the flower appears cyan in the final rendition. And pushing up the red gives those pink tints to the background fence.

 

Oddly, when in Iridient the DNG Temp was given as 1667. But when I white clicked the temperature shot up past 15000. So there are two interesting and very conflicting results from the same app!!

 

As to why there is no green to be brought out in this shot's leaves? I guess green or UV-recorded-in-green-channel just was not present in that particular day's light like it was in the preceding version of the Hibiscus where the green came out nicely in those leaves.

 

Well, those are my best guesses!! Hope this helps shed some light -- some UVIR light -- on everything.

Link to comment

Thanks very much Andrea

Thanks too, for confirming that it was not 'operator error' on my behalf, that couldn't shift the color temp any further........just one of those quirks of using a converted VIS light camera for out of VIS spectrum photography ;)

 

Cheers

Col

Link to comment

I think that pretty much says it all. We are using this equipment outside of its specs, so I suppose we are "lucky" to get what we get. ;)

 

But remember that there are other colour editing tools which can bend the false colours to your will. White balance is a global adjustment affecting all colours simultaneously. But if you use the colour patches in Photo Ninja, for example, you can tweak false colours individually and bring out a desired false-colour look.

 

I think one of the tricky points with all this is deciding what should be "neutral" in an Ultraviolet foto. Even though we have seen a fairly stable white balance response across different cameras under a particular filters, that is just an accident of the Bayer/CMOS technology we are using. So to say that there is a desireable, natural look to UV fotos is probably overstating the case.

Link to comment
I tend to think of "neutral" as something presumed to have a fairly flat reflectance curve in the image passband, i.e. it does not reflect one wavelength more than another. Your PTFE targets are an example of this. So are numerous other things, such as aluminum foil and many other bare metal surfaces, or snow. Things with a flat diffusion curve also suffice--I have used clouds in the sky on an overcast day as a reference, with decent results (I kid you not!)
Link to comment

Yes, you are right!

I seemed to be thinking about all those flower fotos when I made that statement about what to consider neutral in a UV foto. Even though UV-absorbing (moderately- to very-dark) areas on a flower do not necessarily reflect evenly, those areas usually work well enough for a white-balance click if better methods are not available.

Ideally, however, we should include some of those items you mention. I haven't tried clouds! I have used asphalt pavement. ;)

Link to comment

This is what I have been working towards for the past year.

I have even changed from Sigma Foveon to Bayer CFA to get this.

Perhaps I haven't processed this to its best, but only in PhotoNinja, but you'll get the idea.

I am still getting some pink Hibiscus' from this bush

Cheers

Col

 

post-31-0-92853700-1434774520.jpg

 

Pink Hibiscus Flower (to us humans, at least)

Link to comment

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...