spectral transmission function of green filters

[spacediver]

Hi all,

first post here, hope this is an appropriate subforum to post this question in.

I'm currently working on a side project that involves using slow shutter speeds to capture RAW images of gray test patterns on a display, in order to measure very low luminances.

Here is a description of the approach, and some pilot data (using a borrowed DSLR) on this page.

This approach will only be feasible if the white balance of the display is excellent across all measured test patterns, or if the channel that I'm using (the green one) is filtered with something very close to the spectral luminosity function. If at least one of these conditions is not met, there is no guarantee that the RAW data will be proportional to actual luminance.

I've gone to considerable lengths to achieve a very stable white point balance across the grayscale of my display, but it's likely not perfect at the very low end (which is below 0.002 cd/m2). As such, it would be nice to know whether the green filters on most bayer arrays do indeed match the spectral luminosity function, as this will aid me in any purchasing decisions.

Does anyone have any idea about this? Is there a standard that is constant across all bayer arrays, or do they differ from manufacturer to manufacturer? I imagine the info is probably proprietary but figured it couldn't hurt to ask

Any help would be much appreciated, and I'd be happy to clarify anything if my post is too obscure.

[MarkM]

There is a database of camera spectral sensitivity hosted at RIT that might be helpful:

http://www.cis.rit.edu/jwgu/research/camspec/

[deejjjaaaa]

I'm currently working on a side project that involves using slow shutter speeds to capture RAW images of gray test patterns on a display, in order to measure very low luminances.

just buy a camera with CFA washed off sensor...

[spacediver]

There is a database of camera spectral sensitivity hosted at RIT that might be helpful:

http://www.cis.rit.edu/jwgu/research/camspec/

fantastic, thanks for this!

just buy a camera with CFA washed off sensor...

that was my first impulse (or to manually debayer the sensor) but unless I'm mistaken, this would only be useful if one were interested in measuring radiance. Luminance is a spectrally weighted quantity. So either your filter has to match this spectral weighting, or the shape of the spectral signature of your signal (i.e. the relative amount from each wavelength) has to be consistent across the grayscale.

Two different spectral distributions can have identical radiance, but wildly varying luminance. A debayered sensor will not discriminate across wavelength, so will only be proportional to radiance (great for astrophotography, bad for display calibration purposes).

[deejjjaaaa]

that was my first impulse (or to manually debayer the sensor) but unless I'm mistaken, this would only be useful if one were interested in measuring radiance. Luminance is a spectrally weighted quantity. So either your filter has to match this spectral weighting, or the shape of the spectral signature of your signal (i.e. the relative amount from each wavelength) has to be consistent across the grayscale.

Two different spectral distributions can have identical radiance, but wildly varying luminance. A debayered sensor will not discriminate across wavelength, so will only be proportional to radiance (great for astrophotography, bad for display calibration purposes).

make 3 shots with 3 filters on lenses

[spacediver]

make 3 shots with 3 filters on lenses

I'd only need one filter, but that's actually a great idea. I'd get full use of all photosites with whatever filter I wanted then, and I could even choose a filter that matched the spectral luminosity function. Will look into this, thanks.