Hello, I need some help getting started observing. I have done astro-photography for many years at a basic level. I am using a QHY168 cooled CMOS one shot color camera. My work flow is pretty basic: calibrate, de-bayer, align and stack. I have never extracted color channels. I have read (and re-read) the DSLR and CMOS guides but I’m confused about the process. When do I extract the color channels and what do I get when I do?
Do I extract them from my RAW light frames at the beginning, sometime during processing or after I have aligned and stacked them? Do I get three separate files from each light frame? If I have separate color files, do I still de-bayer them?
Help greatly appreciated. EWR
I’ll leave the specifics of your question to someone else with more expertise since I’m also fairly new to this and am working on learning mono CMOS photometry, so only know the DSLR work flow theoretically.
What I will suggest is that I would highly encourage you to reach out to the AAVSO and ask to be paired with a Mentor.
Those manuals can be very daunting at first and mine has been highly helpful.
1 Like
Hello and welcome, @emmett51!
While I’m a spectroscopist and not a photometrist, I think I understand the question. Is it correct to say that this is your current workflow?:
- Calibrate (applying dark(s), bias(es), and flat(s))
- Debayer (resulting in a series of full-color images)
- Register (align)
- Stack
If so, then I think that you can extract the color channels at any point after debayering. To minimize the number of files which you have to deal with, it may make sense to do it at the end, after stacking.
(An aside about stacking: it’s good to note that while it often helps SNR, it’s not required for photometry. It’s also ok to use single images; or, you can record a set, measure each individual image in the set, then average the measurements to obtain a final magnitude. The advantage of doing it that way is that you can also compute the standard deviation to report as error. But that’s not required, either.)
The exact “how & when” of color channel extraction doesn’t matter as much as the end result—you want to be able to say, with absolute confidence, “All of the light in this photometric measurement came only from the pixels with green filters on them.” (Or “blue filters”, or “red filters”—but only one filter at a time.) Often, the easiest way to do that is by selecting those pixels and splitting them into their own separate file, which looks monochrome and which can have photometry done on it just like an image from a monochrome camera. Hopefully, some others will be able to share the details of how they accomplish this.
Also, I second what Ken said:
I’ve heard wonderful things about the mentorship program—make use of it if you can!
As a further resource to check out, you might like to watch Dr. Barbara Harris’ 2021 presentation on DSLR photometry; she covers debayering and color channel extraction starting at around the 30-minute mark:
There should be some more discussion of color CMOS photometry in the AAVSO webinar series archive, on their YouTube channel.
(P.S.: I took the liberty of editing the title of this thread to increase the likelihood that it will be seen by those who can help. Feel free to make further edits if you’d like!)
Thanks for the helpful replies. I have submitted a mentor request and am looking forward to a response. As to my confusion over the initial workflow, I re-read the DSLR guidance a few more times and think I have found the answer (as usual it was there all along).
I understand color extraction can be done either before or after calibration. If done before calibration, the dark, bias and flat frames have to have colors extracted as well. If calibration is done first, the darks, bias and flats do not need to color extracted so fewer files are involved and the whole process is simpler.
After color extraction there should be three grey scale files (RG&B) from each light file. Assuming only the green channel is relevant, the green files can be aligned and stacked to improve SNR or analyzed individually. There doesn’t seem to be any mention of de-bayering in the DSLR guide so I assume its not involved.
My processing software (Nebulosity4) combines the two green channels during batch processing. If not batch processed, the two green channels can be extracted separately. I understand batch processing is acceptable for DSLR work.
This is my planned workflow to start. Suggestions always appreciated. Again, thanks to those who responded to my post.
EWR
1 Like