We're developing several approaches to calibrating your video spectrometer.

Compact Fluorescent bulbs

Feel free to reorganize this text, i just copied it from peoples' emails:

(copied from Dave Haffner's email)

The CFL spectra seems like a good bet for doing spectral calibration of some sort. I think we just need to standardize the make and model. Looks like there is some significant differences in the spectrum of just a few bulks this person bought at the store http://web.ncf.ca/jim/misc/cfl/

It would be worth a survey to find a bulb model that is useful for this task spectra-wise, and one the manufacturer can share a decent spectrum for in their specs. Of course cheap and widely available is important too.

It also occured to me that once we find a good bulb, we could that THE BULB into a lab and measure it's spectrum precisely. Then this spectrum could be used as a reference to calibrate the instruments.

We can do the wavelength registration with a reference spectrum because we just need to match up the peaks, and perhaps use just one channel at a time to do something quantitative with the intensity of the spectra. But to get the benefit of a wider spectral range (as wide as the camera permits) all at once, I think we need to figure out how to combine the bands, at least to the first order. Getting a known bulb into the lab would be a good way to get a sense of how to mix the channels. We could also use sunlight, I think.

(Monty Montgomery contributes:)

I would think the easiest calibrations (the one everyone has access to without buying anything) would be to calibrate the nm scale from one of the two lower mercury lines that are same on every CFL bulb and the 'D' line off a low pressure sodium streetlamp. From there, intensity across the spectrum could be calibrated off ~ any tungsten halogen lamp which follows black body close enough for government work.

Developing a spectral library of calibrated bulb spectra

• marking/tagging some as "calibrated" if they've been tested on other equipment?