![]() ...
·
![]()
·
1
like
|
---|
Here's something that's news to me: https://youtu.be/JLu3Civ2f3Y?si=Vm00BQike482caXQ&t=205 Wish I'd seen this 2 years ago. Has Adam Block covered this before? |
![]() ...
·
![]() |
---|
I'm not sure if Adam Block covered it but if you are using WBPP you just have to check the drizzle box in the Post-Calibration tab.
|
![]() ...
·
![]() |
---|
Adam Block has certainly covered that. As Jung said, its just a check box to enable drizzle in WBPP, Though, you have to have dithered your frames sufficiently during acquisition for that to work properly anyway… |
![]() ...
·
![]() |
---|
What does drizzled ‘sufficiently’ mean? A few pixels every sub-frame?
|
![]() ...
·
![]()
·
1
like
|
---|
What does drizzled ‘sufficiently’ mean? A few pixels every sub-frame? *Any fraction of pixel, below or above unity. Not that using or NOT using SPCC makes much of a difference anyways... |
![]() ...
·
![]() |
---|
andrea tasselli: Here's a comparison of SPCC vs PCC: https://youtu.be/PY6ngRmztzk?si=LB4YBSLKO4YZ4dz1&t=316 Short answer is that SPCC is 400% more precise on average. |
![]() ...
·
![]()
·
2
likes
|
---|
![]() ...
·
![]()
·
1
like
|
---|
John Stone: Isn't dither plus CFA drizzle generally the best way forward for processing an OSC image irrespective of which colour calibration tool is used? |
![]() ...
·
![]()
·
1
like
|
---|
Hello. This is something we cover in the Siril documentation: https://siril.readthedocs.io/en/stable/processing/color-calibration/spcc.html ![]() |
![]() ...
·
![]()
·
1
like
|
---|
Tim Hawkes: *As far as stellar colours go, yes. But you pay that with reduced SNR for distributed objects. |
![]() ...
·
![]() |
---|
What does drizzled ‘sufficiently’ mean? A few pixels every sub-frame? By sufficiently, I was referring to a decent spread of dithers, as well as at least moving the stars a pixel or two on the imaging cameras sensor. Drizzle is a weighted operation (which you can see by looking at the drizzle weighting map if you've got WBPP set up to produce it). Lets say you have 60 subs, with a dither operation between every subexposure, you will have a far better drizzle result than if you have 100 subs dithered every 3 subframes. Drizzle is lossy, the more subs you feed it with more dither operations, the better the outcome will be. But essentially, with an OSC camera, if you want to run spcc, you need to have drizzled the data, which will cost you in terms of final SNR, so you're going to want more integration time than you would otherwise collect. And if you want to drizzle you'll want to be dithering by a reasonable pixel distance, and have a minimum (in my opinion) of 60 dithers in order to produce a good image. If you're dithering every 3rd sub, then, by my reckoning, you want 180 sub exposures to drizzle cleanly. If you're dithering every frame, 60 subs will be enough for the weighting to work out reasonably well, but you may well run into a situation where your SNR is too heavily negatively affected. I guess it comes down to 'For the best possible images with an OSC camera, you want a boatload of integration, lots of dithering operations throughout the image acquisition, drizzle (1x is fine, 2x if you're undersampled, if you're so heavily undersampled that you need greater than a 2x drizzle to reach ~1a-s/p, you should rethink your imaging train) and run SPCC (IF you think its necessary... As Andrea said, it really doesn't make THAT much difference unless your colour balance is absolutely shot!) |
![]() ...
·
![]() |
---|
Wow - thanks, Alex, for the detailed explanation! Neil |
![]() ...
·
![]() |
---|
Tim Hawkes:John Stone: Yes, using CFA drizzle has been recommended for color cameras for quite a while - https://pixinsight.com/forum/index.php?threads/bayer-drizzle-instead-of-de-bayering-with-osc.12996/#post-80658 |