Luminous Landscape Forum
Raw & Post Processing, Printing => Colour Management => Topic started by: BradSmith on April 13, 2015, 02:38:14 am
-
Is there a reasonably close/repeatable method of determining the color temp of a light source using a digital camera and Lightroom and/or Photoshop? I'm not expecting super scientific accuracy. I'd like to be able to check the current color temp of various bulbs and light sources that I have and also check the general accuracy and consistencies of bulbs I purchase in the future. I do not have a color meter.
thanks
Brad
-
Is there a reasonably close/repeatable method of determining the color temp of a light source using a digital camera and Lightroom and/or Photoshop? I'm not expecting super scientific accuracy. I'd like to be able to check the current color temp of various bulbs and light sources that I have and also check the general accuracy and consistencies of bulbs I purchase in the future. I do not have a color meter.
Hi Brad,
Since we're dealing with relative differences here, I suggest to use a light diffuser, I e.g. use a piece of Opal glass for such things, and shoot a direct image of the light source through that. Aiming it at the lightsource will reduce the influence of other ambient light getting into the mix. Exposure time would just follow the camera's exposure meter plus 2 stops exposure correction (to reduce the amount of photon shot noise). Alternatively you could take a shot of a WhiBal card, which has a known, relatively flat, and thus spectrally neutral reflection. The influence of ambient light in reflected light readings is potentially higher than from a direct lightsource measurement. Greycards for exposure determination may not be spectrally neutral.
You then use LR or another Editor to do a Click White Balance, or even an automatic one, as long as you use the same file conversion settings in the future, results should be repeatable.
The determination of Color temperature and Tint can give different values between applications, because there are different methods to calculate the Correlated Color Temperature (CCT). But within the same application it should be usable enough, assuming the illuminant/lightsource is a somewhat continuous spectrum kind of emitter.
Cheers,
Bart
-
Shoot a RAW file of a white card, lit with only the light source to be tested. Do sample dropper white balance adjustment in LR for each frame, color temperature is shown instantly. Can not think of any simpler or more accurate method, if you have no color meter.
-
thanks guys. I had thought of that, but I didn't know if the WB eyedropper tool in Lightroom would show values that were, in fact, the color temp of the source. For whatever it is worth, I've just shot the WB portion of a colorchecker card in a dark room, under a few different bulbs. Here is what I found using the WB tool in Lightroom. Values are Temp, then Tint.
Ott Light 5050, + 53
75W, Par 30 Halogen Flood 2850, +4
Solux MR16 3500K 3000, +11
Same Solux bulb, about 1/2 dimmed 2650, +8
Interesting that the Solux bulbs are so far away from their spec'd color temp.
thanks again
Brad
-
Interesting that the Solux bulbs are so far away from their spec'd color temp.
thanks again
Brad
You did say your procedure isn't scientifically accurate. I've found Solux and any other tungsten based "hot light" to be quite close to their stated color temp just not split hair accurate using your method in ACR/LR.
Not surprised by Ottlight's tint +53 which indicates its green spike. Note other hot lights don't have this severe a shift to magenta.
The Soraa full spectrum 5000K LED flood light I just purchased gives a tint of around -6 to -9 with 4650K to 4850K color temp. And talk about vivid color! Wish this bulb came with a saturation adjustment.
-
thanks guys. I had thought of that, but I didn't know if the WB eyedropper tool in Lightroom would show values that were, in fact, the color temp of the source.
If you take a picture of a white/neutral object, what else the value could be?
-
You did say your procedure isn't scientifically accurate. I've found Solux and any other tungsten based "hot light" to be quite close to their stated color temp just not split hair accurate using your method in ACR/LR.
And yet, when measured using a spectro, not the case, at least with my CCT 4700K lamps:
(http://www.digitaldog.net/files/Solux4700K.jpg)
Nearly CCT 400K ‘off’.
-
thanks guys. I had thought of that, but I didn't know if the WB eyedropper tool in Lightroom would show values that were, in fact, the color temp of the source. For whatever it is worth, I've just shot the WB portion of a colorchecker card in a dark room, under a few different bulbs. Here is what I found using the WB tool in Lightroom. Values are Temp, then Tint.
Ott Light 5050, + 53
75W, Par 30 Halogen Flood 2850, +4
Solux MR16 3500K 3000, +11
Same Solux bulb, about 1/2 dimmed 2650, +8
Interesting that the Solux bulbs are so far away from their spec'd color temp.
thanks again
Brad
the solux color temperatuur is quite ok when voltage at bulba connectors is 12 volt. Normal halogen power supplies are at the terminals around 11.5volt. So at bulb connectors even lower. That is why i use a Adjust able 12v power supply.
-
The Soraa full spectrum 5000K LED flood light I just purchased gives a tint of around -6 to -9 with 4650K to 4850K color temp. And talk about vivid color! Wish this bulb came with a saturation adjustment.
Hi Tim,
Thanks for mentioning the Soraa bulb. I had never heard of it before. Which of their bulb series did you get, the "Vivid" or the "Brilliant". From what you said about saturation, sounds like you might have the "Vivid".
Brad
-
Hi Tim,
Thanks for mentioning the Soraa bulb. I had never heard of it before. Which of their bulb series did you get, the "Vivid" or the "Brilliant". From what you said about saturation, sounds like you might have the "Vivid".
Brad
The vivid series has a higher CRI (color rendering index) as documented here (http://www.soraa.com/public/docs/Spec-Sheets-GU5.3-US/3.0/Soraa%20SM16%207W.pdf). The 36 degree MR16 could be a replacement for the Solux equivalent and could be useful in viewing/proofing prints and has the advantage of being dimmable so that one can adjust the brightness of the viewed print is equivalent to that of the screen. This is important for proofing.
One question: does the color temperature change with dimming as it does with the Solux?
Bill
-
Hi Tim,
Thanks for mentioning the Soraa bulb. I had never heard of it before. Which of their bulb series did you get, the "Vivid" or the "Brilliant". From what you said about saturation, sounds like you might have the "Vivid".
Brad
I bought it off Amazon which has it titled: LED - 18.5 Watt - PAR30 - Long Neck - 100W Equal - 2620 Candlepower - 36 Deg. Flood - 5000K Stark White - Soraa 00807
It is the Vivid Color variety listed here under product code 00807: http://www.soraa.com/products/specs-PAR30L-US
This news article off Soraa's site explains how they achieve CRI 95:
http://www.soraa.com/news/soraa-November-25-2014
I didn't quite understand the flood light dispersion characteristics of this bulb and found it needs to be placed at a distance to use it as a print viewing light or else you get overly vibrant results. I have it stationed 15ft across my living room lighting my 8x10 Epson inkjet prints. Pretty spot on color and a lot of light for one bulb. I shot the example below hand held at 1/8's, f/4.5, ISO 200 where normally with other LED and CFL required ISO 800.
-
One question: does the color temperature change with dimming as it does with the Solux?
Bill
Wish I could tell you but I don't know since I don't have a lamp that allows dimming.
I've been shooting a series one after another self portraits (selfies) standing six feet from the Soraa and can edit one, save as preset and apply to the rest with very little tweaking mainly in green/magenta tint adjusts in ACR. I even compared it to shots taken with my camera flash with (in-camera flash WB preset) and I get almost identical skin rendering. As usual with my face's skin condition non-uniform red blotches still show up but shooting my hands or arms looks very accurate.
-
Below is a comparison to how close the Soraa renders skin to actual daylight.
They're shot Raw and edited to look exactly as they appear which wasn't that hard to do with both.
-
And yet, when measured using a spectro, not the case, at least with my CCT 4700K lamps: Nearly CCT 400K ‘off’.
That's odd. When I measured my Solux, using the same 4700K lamps, I got 4828K. It passed Babelcolor's ISO 3664-test without any problem. And when I checked the spectral data, it showed a very smooth graph. I noticed that you aimed for 2000 lux and I used the “practical appraisal” settings. But this can't change the results so much, or can it?
-
That's odd. When I measured my Solux, using the same 4700K lamps, I got 4828K.
Using what instrument (they are not necessarily going to provide the same readings). I used an i1Pro-2 Spectrophotometer. What lamp housing? This is with their Task Lamp (discontinued).
What this tells us is again, the numbers are not something we can take to the bank. I’d be absolutely shocked if 5 people with differing instruments measured five Solux 4700K lamps and each got exactly 4700K.
-
Using what instrument (they are not necessarily going to provide the same readings). I used an i1Pro-2 Spectrophotometer. What lamp housing? This is with their Task Lamp (discontinued).
What this tells us is again, the numbers are not something we can take to the bank. I’d be absolutely shocked if 5 people with differing instruments measured five Solux 4700K lamps and each got exactly 4700K.
FYI, my Solux 4700 with the Task Lamp shows 4329K. Perhaps the transformer gives lower voltage?
-
FYI, my Solux 4700 with the Task Lamp shows 4329K. Perhaps the transformer gives lower voltage?
Both our values are quite close, perhaps.
-
Using what instrument (they are not necessarily going to provide the same readings). I used an i1Pro-2 Spectrophotometer. What lamp housing? This is with their Task Lamp (discontinued).
What this tells us is again, the numbers are not something we can take to the bank. I’d be absolutely shocked if 5 people with differing instruments measured five Solux 4700K lamps and each got exactly 4700K.
I used the same Task Lamp and measured with an i1Pro2. I used the bulb without black backing. The only difference I can see is that here in Sweden we use 230 V.
I was asked by another manufacturer to see what voltage was required to get the correct color temperature. I saw that when you feed the the Solux lamp with 13,5 V it give a color temperature of 4900K.
-
I guess that if I photograph a white paper illuminated by the light source, it must be possible from the RGB values to find out the color temperature. Anyone knows how to do that?
-
I guess that if I photograph a white paper illuminated by the light source, it must be possible from the RGB values to find out the color temperature. Anyone knows how to do that?
Problem: if the white balance is wrong (in camera or in RAW converter), the white color can be just about anything. So how can you know if the color comes from the illumination or the JPEG/TIFF conversion?
The only solution I can think of is shooting RAW and doing color picker white balance adjustment from the white paper. At least Lightroom provides the color temperature when doing this.
-
I guess that if I photograph a white paper illuminated by the light source, it must be possible from the RGB values to find out the color temperature. Anyone knows how to do that?
Just to find a paper that is neutral white isn't easy. Most ”white“ papers are blueish. And if you did find a neutral white paper and shoot in RAW and open that file in three different programs, you will get three different colour temperature measurements. So it will depend on how accurate you have to be.
-
In my experience I've been able to tweak ACR/LR color temp sliders on a Raw shot of a flat spectra WhiBal card after clicking for R=G=B and not change the numbers but noticeably change the color cast of the scene so there goes your accuracy and repeatability.
Color temp data is the only manufacturer info as tags in the Raw's metadata that third party converters can use with regard to controlling color appearance but it's still interpreted.
Spectra data that describes exact luminance scaling of HSL appearance of backlit RGGB filtering on the sensor as more light is exposed to the sensor is still somewhat guess work as to how it should be appear reconstructed on a computer video system.
Think of the behavior of how stain glass window inside a church appears or changes hue/sat viewing at daylight, late afternoon and at night with car headlamps back lighting it. Does the green glass at night look like the same green when it's backlit by more light? Does it scale linearly? No one has proven what this looks like and how much it affects the hue/sat of color temp appearance. We have to rely on reconstruction on a limited gamut display.
Finally someone posted online a photo of an actual back lit RGGB bayer filter to demonstrate whether these RGB purities scale linearly exposed to more light as in the stain glass example...
http://www.siliconimaging.com/RGB%20Bayer.htm