Is there a (simple) way to get at a particular camera sensors spectral sensitivities?
The methods mentioned (CCSG, grating etc.) all face the calibration problem or provide a problem with the accuracy and a monochromator is everything but simple. There are a few devices like camSPECS designed for a quick and easy measurement and they are cheaper than a monochromator but not that cheap. The easiest and cheapest way to get spectral sensitivities for a single device is to have it measured as a service. Dietmar
On Mar 16, 2016, at 9:36 AM, Wüller Dietmar <dwueller@gmx.de> wrote:
The easiest and cheapest way to get spectral sensitivities for a single device is to have it measured as a service.
Easiest, maybe, but certainly not cheapest. And perhaps not best, either. The bill of materials for a suitable spectroscope is under $20. You'll need to know the transmission efficiency of the diffraction grating, and the least-worst way to get that is with a monochromator...but you can get a monochromator for cheap off FleaBay, and a spectroscope design can be adapted to a monochromator design if you want to continue the dirt-cheap route. And virtually nobody I'm aware of (with one notable exception who's reading these words (Hi Iliah!)) takes the lens into consideration when determining a camera's spectral sensitivities. Some lenses have rather complex spectral absorption characteristics. If you must use a lens, use Canon's "Plastic Fantastic" 50mm f/1.8; its transmission is as spectrally flat as you'll find in any lens and significantly flatter than many, including lots of top-of-the-line L glass. But if your commercial service (or whatever) assumes that the optics are spectrally flat, you've got the lens permanently "baked into" all your data. Cheers, b&
participants (2)
-
Ben Goren
-
Wüller Dietmar