Jump to content
UltravioletPhotography

GH1 Broadband Profiling Examples


Andrea B.

Recommended Posts

This morning I was playing with some old Lumix GH1 UV photos of standards and bullseye flowers. I wanted to look into colour profiling the GH1 because I hadn't really done a good job the first time through.

 

My broadband Lumix GH1 is wonderful at setting an in-camera white balance through most UV or IR filters - as are its cousins in the Lumix G line. In-camera white balance through dark filters cannot be done as easily, if at all, with Nikon broadband conversions.

 

In these fotos I was using an early test version of the Andrea-U on some lens which I unfortunately don't remember. All photos were brightened somewhat to overcome the narrow dynamic range of the GH1.

 

Baseline: GH1, Andrea-U, In-camera Preset White Balance

Nothing at all wrong with this as a UV photograph. The flowers' UV-dark bullseyes are looking good!

GH1_AndreaUV_wbManual_f8ss1-8iso100_20110813wf_1336260_inCamWBPreset_NX2Brighten.jpg

 

*****

 

In Photo Ninja I made a visible Custom Light profile named GH1-Vis from the Mini Colour Checker card as photographed in a visible foto (not shown). I then used GH1-Vis as a Light Source choice in the Correction Tool and applied it to the UV photo. White balance was measured in the UV photo and saved along with the Light Source setting as a UV Preset for future use.

 

Profiled Colour: GH1, Andrea-U, Light Source = GH1-Vis, WB = Manual from white standard

This does shift the colours, of course, because the missing internal filtration has altered the GH1 colour profile.

GH1_AndreaUV_wbManual_f8ss1-8iso100_20110813wf_1336260_inCamWBPreset_PNCCProfileBrighten.jpg

 

*****

 

Photo Ninja has a built-in, standardized colour profile named Daylight/Flash. It also has the Match Color Temp choice. It turned out that Daylight/Flash and Match Color Temp gave almost an identical histogram for this particular photograph, although I couldn't tell you why. Note that these two photos are really not too far off the Profiled Colour version. How much this might hold true for other photographs, who knows?

 

PN Built-In #1: GH1, Andrea-U, Light Source = Daylight/Flash, WB = Manual as saved from white standard

The colour patches on the Mini CC are not as violet as the Baseline version and not as blue as the Profiled version.

GH1_AndreaUV_wbManual_f8ss1-8iso100_20110813wf_1336260_inCamWBPreset_PNDaylightBrighten.jpg

 

PN Built-In #2: GH1, Andrea-U, Light Source = Match Color Temp, WB = Manual as saved from white standard

GH1_AndreaUV_wbManual_f8ss1-8iso100_20110813wf_1336260_inCamWBPreset_PNMatchColTempBrighten.jpg

 

*****

 

PN No Profile: Gh1, Andrea-U, Light Source = No Profile, WB = Manual as saved from white standard

I have no idea where the colours are really coming from with this Light Source choice, but I don't like them!! :blink:

GH1_AndreaUV_wbManual_f8ss1-8iso100_20110813wf_1336260_inCamWBPreset_PNNoProfBrighten.jpg

Link to comment

Using the B+W 092 IR-Pass filter, I made an in-camera white balance on the white standards which gave a very pretty rendition. This filter passes some red light, so you usually get blue tones after any wb step.

GH1_092IR_wbManual_f8ss1-320iso100_20110813wf_134231.jpg

 

 

Here I've applied the Photo Ninja GH1-Vis colour profile as the Light Source choice in the Correction Tool. The white balance was re-measured on the white standards. There is some change in the colours, as expected. But I'm not sure I prefer this over the in-camera wb. Do you?

GH1_092IR_wbManual_f8ss1-320iso100_20110813wf_134231_PNProfile.jpg

Link to comment

This experiment with the U340 UV-Pass filter need to be repeated because I'm not sure why I could not get a good in-camera white balance. The U340 was stacked with an S8612 to block IR. Everything is yellow!

GH1_340UV_wbManual_f8ss2.5iso100_20110813wf_134053_nx2.jpg

 

After white balancing on the white standards and applying the GH1-Vis profile to the colours, I got this noisy mess. The white balance step clearly did not carry over to the black standard which looks somewhat blue. However, this is another reason why I need to repeat a test with the U340 because it is possible that the colours on the periphery of the photo were affected by the stacking of two dark filters which could cause vignetting or colour change.

GH1_340UV_wbHalogen_f8ss1.6iso100_20110813wf_133029_PNCCProfile.jpg

 

Here is a 100% extract of the black standard so that you can see the dark blue which is not so obvious in the preceding resized jpg. Maybe it is just blue noise?

GH1_340UV_wbHalogen_f8ss1.6iso100_20110813wf_133029_PNCCProfilepnCrop.jpg

Link to comment
There is no guarantee you can get 'UV-white' that looks white in a narrow spectral range. For this to occur, all three channels must provide the (near) identical read-out. I also get the same dirty-looking yellow rendition in UV with my U340 using any of the metrhods for setting white balance. That simply means (almost) no response in the blue channel for this narrow band. If you force the image to give white, horrendous noise in blue will result. Exactly as you have evidenced.
Link to comment

Does the U340 have such narrow a range? It goes from 200 to 350nm or so.

 

I would like to try the U340 on the D600 to see if it has less noise.

 

The GH1 shot was for 1.6" @ ISO-100. Quite a lot of noise for base ISO of 100 it seems. But wrenching the white balance around always induces noise even at base ISOs, as you have observed.

Link to comment
Although the filter (U340) appears to cover a wide range, most of the transmittance is peaked around 340 nm. Whether you can w/b against a UV-neutral object directly in camera (Panasonics), or do the w/b later (Nikons etc.), the result is the same with the various cameras I've tried using this filter on a UV-Nikkor. The exception being the Fuji S3 UV/IR LE, as this model has almost no response at 340 nm, so needs massive UV exposure to get anything recorded at all. However, again that "something" is rendered dirty yellow (with a awful noise component).
Link to comment

The problem, it seems to me, with your U340 test is the stacking of the S8612. That crunches the FWHM and right-shifts the peak tremendously. Below is a jpg using a U340 within a CopperU cell. I used the GF1 in-camera WB. No PP except for reducing image size from 4k to 1.5k px width.

Panasonic Lumix GF1, ISO 400, CZJ Tessar 50/3.5, f8.0, 0.625s, halogen lighting.

 

post-19-0-15418300-1425566557.jpg

 

And below is the same after one-click WB and reduced to 1.5k width.

 

post-19-0-04475300-1425566782.jpg

 

and a 100% crop of 4k with the one-click WB

 

post-19-0-31461200-1425567044.jpg

Link to comment

Andrea,

I also encountered this failure to get a good in camera UV-white balance with my G3 some while ago. I was using a Baader-U under lighting from a ReptiSun 10.0 UVB T5 HO - High Output Linear Lamp (Zoo Med Laboratories, Inc). The in camera WB on a Spectralon target was yellow like yours above. Unlike your experience, I was able to recover what I thought was an acceptable WB in Photo Ninja using one of the built in profiles, most likely Match Color Temp. I did not however photograph the ColorChecker, only the critter.

Link to comment

John, You have a photo of a reptile in UV? May we see it ??

 

**********

 

On the wb thing -- repeat the in-camera wb effort several times from slightly different angles and eventually you sometimes manage to 'catch' a good one. Although sometimes not.

 

**********

 

Reed, stacking could indeed be a culprit in the wb problem. :D

Link to comment

I was evaluating the reptile lamp for use in UV-photography and to access how reptiles and amphibians UV-vision might be supported by the lamp.

 

added later: Here is info on the ReptiSun 10.0 UVB T5 HO - High Output Linear Lamp (Zoo Med Laboratories, Inc)

 

The subject in my photos was one of my patio toads. I am planning to post a formal UV Fauna of the toad, but will post the yellow UV-WB shots here since it relates to your similar observation above.

 

The files are on one of my PCs at work but we were iced in again Wednesday night here in west Tennessee. I am telecommuting for the 2nd day, so I won't get to the files today.

 

The TV weather report said we have broken cold temperature records from sometime back in the 1920's!!

Where is that global warming when you need it?

Link to comment

Warming of any type is certainly missing here too. We just got 7 more inches of snow. Enough already!

 

You have toads!! So cool. We have some here in NJ and in Maine but I've never seen one, just frogs.

Link to comment
Bill De Jager

Where is that global warming when you need it?

 

Out west where we have been having record warmth for the past couple of years, a grinding drought in California, and now record-low snowfall in the Pacific coast states. :angry:

 

Andrea, I'll have to try this same thing on my GX1 which should be good practice!

Link to comment

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...