2x2 binning - one shot color

[dpaul]

I have a fundamental question about imaging processing with a one-shot color CMOS camera. When saving RAW fits files using 2x2 binning, I'm then having problems debayering the results to get any strong color/right color. I'm wondering if this is because of the binning. I haven't tried 1x1 yet but will do so when its next clear one night.

So fundamentally, can Pixinsight support processing of 2x2 binned frames?  This is software binning with CMOS so can the debayering process handle this somehow? It must be screwing up the CFA data and explains why it can't autdetect CFA, I have to do it manually.

I have an unguided scope so it would be really beneficial if I can use 2x2 binning with OSC. Also for my scope its already very over-sampled with 2x2. I always use 2x2 with mono cameras and of course that's not an issue.

If this isn't currently possible with Pixinsight, it would be a great addition if that can be done.

Thanks

David

[pfile]

you say "saving raw fits files using 2x2 binning" but then say "this is software binning".

do you mean that your capture software is binning the data? if it bins 4 adjacent pixels together then you're just getting a "pseudo-L" image - red/green/green/blue pixels have been averaged together to make an "L" pixel. so i don't think there's any color to extract.

i'm pretty sure bin 2 of an OSC does not make sense and people don't do it, but i'm happy to be wrong (say for instance if the software only picks out the red pixels from the matrix to create a red pixel in the bin 2 output).

rob

[dpaul]

Hi Rob,

I believe 2x2 binning with a CMOS puts data from 4 pixels into a single pixel but in 'software' not on the chip like CCD.
When viewing live with my camera, it debayers on the fly and shows color images, so using 2x2 binning I can get more detail with short exposures than with 1x1. However when capturing data for processing, I save FITS files non-debayered so this must of course be affecting the CFA array so its no surprise I get screwed up color when using debayering process in PI.

So my question is ''could Pixinsight be programmed to debayer a captured 2x2 frame - surely that's possible?

Thanks

David

 

[RickS]

So my question is ''could Pixinsight be programmed to debayer a captured 2x2 frame - surely that's possible?

I don't think so.  Once you've added the 4 pixels in each 2x2 super pixel together there's no way to go back and figure out the relative contribution from each colour.

[dpaul]

Ok Rick,

I'll accept defeat on this one!

Appreciate all the responses.

David