Jump to content
UltravioletPhotography

Lens and camera tests with sparticle and monochromator


Jim Lloyd

Recommended Posts

Are you using any color profiles for this system of camera+lens+filter+light?

 

Something which might be useful....take each of your predicted and actual colors and find their color wheel reference color at full sat and brightness. It isn't all that important to do this, but does sometimes give a better "feel" for what a color is. For example your 407 color bars are so dark that I cannot see what color they might be.

 

Here is the same set of actual and predicted colours as above

 

post-175-0-34255700-1535552250.jpg

 

Here i have set lumiance to 125 in all cases (I am using Excel which has luminance values between 1 and 255 - so about 50%)

 

This shows colour predicition is very close.

Link to comment
Thanks for indulging me! That was interesting to see. I do know we won't ever see much of those fully saturated, fully bright colours, if any. But I had no "feel" for those dark versions. :D
Link to comment

Jim, Not sure why your green and yellow look so brown when white balanced. Similar to you, I use very thin PTFE installed in one of the Sparticle holes,

it should be about the same as the tape I guess, never tried the tape that way though.

It still works for what it is for, finding the depth of transmission, the color doesn't matter for testing transmission depth.

I don't see the reason for RAW colors, they are white balanced too, everything has a white balance, but the RAW white balance is just not optimal, but still works pretty good for testing depth I guess.

By the way, I think I have an extra Sparticle holder around here somewhere if you want to pay for shipping, I would give it to you. It is PTFE and I would include the thin PTFE I use for white balance.

Fist Class would probably be about $12? to you, I think.

What size diameter on the outside are your BP filters? 12.5mm diameter?

Link to comment

Not sure why your green and yellow look so brown when white balanced.

 

Our color perception is non-linear. At a certain point, as a color is lightened or darkened, we perceive a hue shift even though there isn't one. The Abney effect.

 

Example: As you desaturate pure blue by adding white, eventually the eye sees "lavender" or some pale purplish hue.

 

When I see yellow becoming darker, it eventually looks olive green to me rather than dark yellow.

 

One of Jim's colors is yellow. The other two have lesser amounts of green. As these colors darken, we see the hue shift.

colorDot.jpg

 

Actually there is a bit of an anomaly there.

The green in the color associated with 342 is 225.

The green in the color associated with 356 is 195.

Then in the color associated with 366, the green increases again to 204.

From top to bottom, the color wheel bumps from 53° to 46° and then back up to 48°.

Weird.

Could just be the effect of stuffing it all into a color space.

There are a lot of reasons why we need to be very very careful with these false color analyses. It is useful to understand color spaces and how they are constructed. I do not wish to again belabor the lack of a 1-1 wavelength-color mapping, but that is not the only snag which needs explaining to make this all work. For example, what is the proper intent to be used in the color space to handle anything which might lie outside the gamut? Oh well.....another day perhaps....I've barely scratched the surface myself in trying to understand color - whether false or non-false.

Link to comment

Jim, Not sure why your green and yellow look so brown when white balanced. Similar to you, I use very thin PTFE installed in one of the Sparticle holes,

it should be about the same as the tape I guess, never tried the tape that way though.

It still works for what it is for, finding the depth of transmission, the color doesn't matter for testing transmission depth.

I don't see the reason for RAW colors, they are white balanced too, everything has a white balance, but the RAW white balance is just not optimal, but still works pretty good for testing depth I guess.

By the way, I think I have an extra Sparticle holder around here somewhere if you want to pay for shipping, I would give it to you. It is PTFE and I would include the thin PTFE I use for white balance.

Fist Class would probably be about $12? to you, I think.

What size diameter on the outside are your BP filters? 12.5mm diameter?

 

Re: RAW - in this case when I am refering to RAW there is no white balance applied. In the sparticle images they are the values given in Rawdigger and displayed as raw composite. In the model predictions they are derived from the camera response and filter transmission.

 

Re the colours - I have spent ages looking at this - basically there are a few factors which interact so its complicated. And there is of course a difference between my prediction and the actual.

 

Regarding my intitial measurements it is likley there is some stray light which has the effect of giving some false response which overall has the effect of reducing saturation in the predictions. Although this effect is small it will get amplified during the gamma correction and white balance process.

 

Having said that, I do find that the clarity of the yellow and green seen in the actual sparticle image can be rather "dirty", but the exact colour percived depends a lot on the software being used and exactly how its done.

 

In the end I am satisfied that the predicted colours and actual colours are close enough to give me confidence that my plots of the sensor spectral response are close to the truth (say within 10%?).

Link to comment
At a certain point, as a color is lightened or darkened, we perceive a hue shift even though there isn't one. The Abney effect. When I see yellow becoming darker, it eventually looks olive green to me rather than dark yellow.

 

I ran into this effect with my mom - we disagreed over the hue in the buoy here. I was pulling for yellow, she declared it was green, and because I have some color blindness, I couldn't argue against it (even though my color picker also said yellow).

 

post-94-0-32569800-1535579017.jpg

Link to comment
and to make matters more complicated we dont percieve colours in isolation, but perception depends on other colours around the object of interest. The RGB receptors (I know that is inaccurate terminology) work to detect R-G and Blue-Yellow(R+G) differences rather than detecting each channel separately.
Link to comment

I don't know why your 340/350/360 range is all brown, not green or yellow like mine, and other's show when optimally white balanced.

Yours are the only examples I have seen from anyone using this method that don't show a green and yellow.

Your lens filter stack 'should' be OK for removing any OOB leak.

Basically, I am saying your brown filters 'should' be green and yellow, should meaning according to results I have seen other people show.

Not sure what the difference is.

Your examples in the posts below shows more of a green/yellow.

#7

http://www.ultravioletphotography.com/content/index.php/topic/2902-lens-and-camera-tests-with-sparticle-and-monochromator/page__view__findpost__p__22874

#38

http://www.ultravioletphotography.com/content/index.php/topic/2902-lens-and-camera-tests-with-sparticle-and-monochromator/page__view__findpost__p__23193

 

post-87-0-03010200-1535592572.jpg

Link to comment

Whether the filters in the range 340-360 look a bit brownish comes down I think to how much response there is in the blue channel.

 

If there is some (even very small) then how this appears will depend on details of software rendering

 

In my case there is a small amount of blue in this region in the RAW sparticle data and in the response measured using the monochromator.

 

So the fundemental question is whether the blue Bayer dyes do transmit some radiation in this range.

 

My feeling is probably yes, although for both my methods of measuring there may be alternative explanation: With the monochromator I only had the BG40 filter as I wanted to see the response in visible (particularly 400-500) so that I could then model the effect of different filters cutting at different points atound 400. If taking the work forward I would definetly look at this more. With the sparticle then there may be some small transmission outside the band of interest and then may be some contribution from reflected light.

 

Having said that I still think there may be some genuine transmission through the blue Bayer. I think Jonathan's work might show this (see here) and somewhere (can't find the link just now) I saw some work by Kluass Schmidt indicating blue response at short wavelengths. Also when I looked before at your (Cadmium) sparticle images - although they do look green and yellow (in this range) - they do have a little blue, even if this isnt very obvious visually.

 

So if there is some genuine small response in blue at these wavelengths how it appears in the final rendered image will be sensitive to processing. It will get boosted considerably by both the white balance (as overall blue reponse is low compared to red) and by gamma correction.

 

I have looked at my images with Rawdigger, View NX2, FastRaw viewer, Photoninja, and Lightroom and all give slightly different results which vary with different settings. They go from quite clear green and yellow, to muddy brown versions. The most recent images above are from View NX2 using the Neutral profile as I thought this would be the minimal level of processing that would be most similar to how I modelled the white balance process.

 

For actual photography I would tend to produce something that looked a little brighter.

Link to comment

Please sign in to comment

You will be able to leave a comment after signing in



Sign In Now
×
×
  • Create New...