Problem with order of processing

[Guest]

I took a series of images of a standard star field in order to calculate the transform coefficients for the photometry. This involves calculating the b-v values for the standard stars but I had a problem which turned out to be the order in which I did the processing. I'm using a DSLR with a telescope and MaximDL software for the processing.

Initially I did it like this -

Calibrate each image

Stack all the calibrated images

Split the stacked image into R, G1, G2 and B bayer layers.

Do the aperture measurements for each colour then calculate the b-v values.

The problem I had was that the b-v values were almost exactly the same even for stars which had big differences in the catalogue values of B-V.  I did a further check by measuring the intensity of a star in the stacked image, then measure the intensity of the same star in each of the R, G1, G2, B images. I found if I added up the four individual colour intensities it was equal to the intensity of the same star in the stacked image. Which is what I expected however it also appeared that the individual colour intensities were almost exactly 1/4 the total intensity. This was the case even for stars which had big differences in the catalogue B-V values. It was as if it had taken the intensity in the stacked images and simply divided by four to get the R, G1,G2 and B intensities.

So then I did the processing like this -

Calibrate each image

Split each image into R, G1, G2 and B components

Stack all the R images, stack all the G1 images, stack all the G2 images, stack all the B images.

Do the photometry.

This time the b-v values came out exactly as I expected. When I plotted the measure b-v values against the catalogue B-V values I got a nice straight line with the right kind of slope.

So I know how to do it now but I'm curious to know why the first method was wrong if anyone has any ideas.

Cheers

Steve

 

[vlaiv]

If you stack your data prior to debayering it will effectively destroy color information and turn your recording into (low quality) mono / luminance information.

This happens because stacking does alignment and different color pixels end up in same place - so you stack red against green against blue - and you effectively average them out - which creates gray color.

When you debayer after that you'll just get gray image where B-V information is constant.

[Guest]

Thanks Vlaiv, I thought you'd be the one to answer this question. Much appreciated.

Steve