15th Feb 2024: Astro Pixel Processor 2.0.0-beta29 released - macOS native File Chooser, macOS CMD-Q fixed, read-only Fits on network fixed and other bug fixes
7th December 2023: added payment option Alipay to purchase Astro Pixel Processor from China, Hong Kong, Macau, Taiwan, Korea, Japan and other countries where Alipay is used.
Hello Mabula,
At first thanx for writing such a great piece of software, and great support on this forum, and I look forward to the development of APP in the future.
I’m used to calibrate and stack my images in DSS, so there is much to learn for me.
I did use APP for some of my old en new data, and got good results, following the DSLR work flow.
But last month I capture some “ problematic” data, trying to process in APP that gives me less good results, mainly due user error I guess.
That data became “ problematic” due the fact that after the meridian flip,
my Canon EOS UV/IR clip-in filter shifted.
(I was fooling around, to find the right spacing for my focal reducer,
and testing some dithering settings,
knowing now my dithering has to be more aggressive in the future)
Anyway, I ended up with some images that has more than normal vignetting.
I did take flats before and after the filter shifted, curious of there was a way to rescue some data.
The stack contains:
1 light frame of 8 minute
5 light frames of 10 minutes
17 light frames of 10 minutes (after meridian flip)
3 dark frames of 8 minutes
3 dark frames of 10 minutes
25 flat frames before meridian flip
25 flat frames after flip
40 bias frames
In the integrate tab, I used the weight option “ Quality”, but maybe this was not a good choice given my situation.
Resulting in the image below:
It seems to me, the vignetted flats did not work here.
After that I followed, another work flow in the Sticky section :
“ Combining subs with multiple exposure time. “
Which was not totally clear to me, I registered the normal en the vignetted data apart from each other, and integrated those files together, but I was getting much error messages about wrong dimension sizes, errors I’ve never seen before.
So it’ s obvious there must be an user error in this case…
Out of curiosity I loaded all the RAW files in DSS and loaded the RAWs in 2 different groups, to see what that program would come up with, and surprisingly, the final stack has less vignetting, it seems somehow the vignetted flats did work here.
Is it possible for you to show me what I did wrong in APP, and how to do the calibration and integrating the right way, step for step?
Beside this problem above it is not totally clear to me how to calibrate and integrate DSLR images of different exposure times.
I know it is easy in DSS, using different groups, but I like to learn how to do this in APP.
Take your time, I know you must be busy….
Thank you for reading!
p.s. I posted this question yesterday, but I think something went wrong, I also missed a preview/edit function, to check the text/pictures before hitting enter..
so I’m trying again with a “compact version” of this question. 😉
Hi,
Mabula has recently done a workflow and video for me with regards to calibration and integration of DSLR images including lights, dark, flats and bias frames.
He informed me that he was going to put the video on his site to help other DSLR users but if you want a copy of the video etc I don't mind sending it over to you to see if it helps.
Regards
Doug
Hi Betelgeuze,
Thank you for your nice compliments 😉
Yes, I see from the screenshots that probably the flat-field calibration is not working like it should.
You need to calibrate the flats with a masterbias and you can use the darks and masterflat to calibrate the lights. And I suggest you create a BPM as well.
The video tutorial that I made for Douglas is perfect for your situation. And you can use the 8 minute darks for the 8 minute light frame, and the 10 minute darks for the 10 minute light frame.
The video is here:
Later today, I will add it to the tutorials on my site as well 😉
Let me know if this helps you in getting good calibration.
Kind regards,
Mabula
Hi,
Mabula has recently done a workflow and video for me with regards to calibration and integration of DSLR images including lights, dark, flats and bias frames.
He informed me that he was going to put the video on his site to help other DSLR users but if you want a copy of the video etc I don't mind sending it over to you to see if it helps.
Regards
Doug
Thank you Doug for assisting, I will add the tutorial to my site later today 😉
Thank you Douglas and Mabula for your help, much appreciated.
I'm gonna check out the tutorial. 🙂
Update:
The tutorial was an eye opener, concerning how to manually make calibration files, and the nice info in between..
(normally I did use the " automatic" way, including making a BPM..)
But... the other eye stayed closed, so to say... it did not solve my problem I' m afraid.. 🙁
I' m totally aware of the fact that this is not a common problem of course, and just a once in a lifetime experience ... 😉
Anyway..I tried different possible and un-possible scenarios, calibrate and register the vignetted images apart or together, with 2 separate master bias files (the heavy vignetted, and the normal), and 1 master bias containing normal en heavy vignetted flats, but I could know that' s never gonna work...
In all cases the final stack or just the check with the " l-calibrated" slider was not satisfying...
But... if I ONLY checked the heavy vignetted images & the heavy vignetted masterflat (+ master dark + BPM), you could clearly see the vignetting went away...
single vignetted :
single vignetted + vignetted masterflat
But the vignetting came back when I involved and checked the normal images & the normal masterflat...
I hope you can still follow me... 😉
So to simplify my situation...
I have:
5 (normal) Lights (each 10 minutes) + 25 normal flats... (created a masterflat)
17 (heavy vignetted) Lights (each 10 minutes) + 25 heavy vignetted flats... (created a masterflat)
3 Darks (each 10 minutes) (created a master dark)
40 Bias (created a masterbias, used only once)
1 BPM
How can these be successfully integrated to 1 final image, without the heavy vignetting.
Maybe I have to treat this data with something of " different optics" option..?
Hi Betelgeuze,
Okay, good to hear that the tutorial was helpfull 😉
So for your normal lights you have an appropriate masterflat and for your heavy vignetted lights you have an appropriate masterflat as well, that's perfect. What I would do in that case:
Calibrate the normal lights with the normal masterflat, masterdark and BPM and save them with the save calibrated frames button.
Clear the frame list panel and do the same for your have vignetted lights with the heavy vignetted masterflat, masterdark and BPM.
Then clear the frame list again. Now load all the calibrated light frames and proceed with star analysis up until integration. ( Check if the calibrated frames that you saved look okay before continuing though. )
That should do the trick I think 😉
Mabula
Excellent !
I did learn a few things the past day's !
Thank you for this Astro Pixel Processor Insight... 😉
So.... I happily ended up much wiser, and with almost no vignetting on the "problem" image, following the suggested work flow.
But when I zoomed in, I saw a few stars that are having a greenish halo...
I integrate and calibrate again with different settings, to see if that helped.
In calibrate with and without the align channels function.
In register I tried, with and without dynamic distortion correction.
In Integrate I used Weights : both Quality and Star shape, Median,
Lnc 1 , 1 and 3 iterations, enable mbb.
Filter sigma clipping, kappa 2.5 , 1 iteration
The results are quite the same.
Then out of curiosity I loaded the DSS fits stack in APP, which seems a bit brighter/colorful, and had the same Tools work flow applied identical values, as the image stacked in APP, and I do not see any green halo's in this image.
Is there an explanation for this, or are there things I can improve, and fine-tune in APP ?
Hi Betelgeuze,
Excellent 😉
The differences that you see are probably caused by one of the 2 separate factors.
- The debayering algorithm that you used in APP and DSS, try different algorithms to see if that gives different results.
- any option for color calibration and/or white balance, I think the color balance in the images shown is a bit different which in itself could explain the difference that you see. (I think this is the most likely reason)
Kind regards,
Mabula
