how to get photometric data from film

When shooting stars with film it should be possible to use known
magnitudes to calibrate film intensity to (better) photometric accuracy.
Thats why I ask in this group,in the hope that this has been done.

Does somebody know how to process a raw scan (Velvia on a Polaroid SS4000) so
that the intensity of the resulting image is photometricaly locally correct?
Lets also say I scan at 4000dpi and downsample it to 1000dpi so that individual
grains want be visible.
Is it possible to get some calibration-file from the manifacturer?
Does a films inherent unlinearity differ much between brands?

I need this calibration for successful deconvolution.

Per Sabelström www.quantimage.com

On 21 Jul 2003 06:27:12 -0700, (per sabelstrom) wrote:

When shooting stars with film it should be possible to use known
magnitudes to calibrate film intensity to (better) photometric accuracy.
Thats why I ask in this group,in the hope that this has been done.

Does somebody know how to process a raw scan (Velvia on a Polaroid SS4000) so
that the intensity of the resulting image is photometricaly locally correct?
Lets also say I scan at 4000dpi and downsample it to 1000dpi so that individual
grains want be visible.
Is it possible to get some calibration-file from the manifacturer?
Does a films inherent unlinearity differ much between brands?

I need this calibration for successful deconvolution.

Getting photometric data from film is very difficult. There are numerous
problems, centered mainly around the film's non-linearity. This, and the exact
color response, vary from emulsion lot to lot, with film age, and with the film
environment (temperature and humidity.) Add to this that your data is on color
film, which is much worse for accuracy and presumably suggests your data wasn't
collected using photometric filters. The low dynamic range of film also greatly
reduces the number of comparison stars available in your field. There are good
reasons that film is no longer used for precise photometry!

To get anything like reasonable results (say, within 0.1 mag) you need what
would normally be considered an underexposed image, so the stars you want to
measure are in the linear range of the film. In typical deep sky astrophotos,
the stars are overexposed. Scan at a fairly high resolution, but you don't need
to be extreme. Film has fairly low resolution compared with a CCD, and the star
images will be fairly large. You don't need to scan individual film grains. You
need to use a scanner that will output the data at 12 to 16 bits per channel.
Depending on the color response of the film, you may be able to somewhat emulate
filtered data by working only with single color channels, eg use the green
channel as a substitute for V filtered raw data.

Save the scanned channel as a 16-bit TIFF and open it in an analysis program
that can handle this format. You can also open it in Photoshop and export it as
a FITS file using the free plugin that is available. Use a standard photometry
tool on your data, but make sure your comparison stars are similar in brightness
to your star under measurement. I'd keep them within 2 magnitudes.

If you already have some data on film that you want to analyze, you may be able
to get decent results. If you are starting a photometry program, you are making
things very difficult for yourself using film. With even an inexpensive CCD you
can get between 10 and 100 times the photometric precision of film.

_________________________________________________

Chris L Peterson
Cloudbait Observatory
http://www.cloudbait.com