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.
I've watched a few tutorials on creating a BPM. I basically need to take long darks (I'm doing 600 second darks), flats, and dark flats. Then I load them all up and tell APP to create the BPM.
Ok. No problem there. But when I go to actually start my work flow to integrate lights, do I just load the BPM as a master? I don't need the darks, flats, or dark flats that I made specifically to create the BPM anymore do I? Can those be deleted? I just need the BPM master for future integration with lights?
I realize I need other lights, darks, and dark flats to match my light frames for integration. I'm just asking if there is any purpose of keeping the subs I used SPECIFICALLY for creating the BPM.
I would keep the dark and dark-flats as those can be used in the future maybe? The BPM will help a lot in the place of darks and can be used for a very long time, but I still would use more recent darks (taken every few months or so) as well as these can contain new bad pixels and ofcourse it still corrects better for things like amp-glow. A combination of all these with the BPM is the best. Neither of them will remove 100% of the noise, it's always a tradeoff and statistical game.
Thanks Vincent. Just so I understand, when I'm integrating lights, I'll use the BPM in addition to regular darks, right? The BPM doesn't replace the needs for darks. It just helps along with darks?
Also, I created two different BPMs using Hot Pixel Kappa 3.0 for one and 2.5 for the other. Both are showing my bad/hot pixels at 8.985 percent. Does this seem high? I thought I read that it needs to be in the 3% range.
when I'm integrating lights, I'll use the BPM in addition to regular darks, right?
Yes, it helps along. If the BPM corrects enough and you don't have any amp-glow, you can check this by comparing both approaches, you might do away with darks.
The percentage of hot pixels is on the high side yes. You can lower the kappa based on what you see happening with calibration. So uncheck the "scale" checkbox in the viewer, zoom into a section of background with hot pixels, load in your BPM with a certain kappa value, change to L-calibrated and see if it works good enough. You can then raise or lower the kappa and see roughly what this value should be for your camera. You could be a bit more aggressive and lower it more.
I've made two BPMs: One using 300 second darks and one using 600 second darks. The 300 second BPM shows a bad pixel percentage at 7.8% while the 600 second BPM shows a bad pixel percentage of 8.9%. Both used a hot pixel kappa of 3.0.
Is this something to be concerned with? I see other tutorials where people are getting a bad pixel percentage of ~3%. Why is mine so much higher?
Well I can only say it may be your camera has a bit too high of an amount of bad pixels? Did you play around with the kappa before creating the BPM like I mentioned above?
Yes. Tried kappa 2 - 3. All kinds reported the same level of bad pixels.
Mm, maybe try between 1-2? If it remains, it must be that there are a lot of bad pixels on the sensor.
So going with a lower hot pixel kappa actually increases the amount of hot pixels reported. A kappa value of 1 increased the hot pixel percentage to 16%!
So I went the other way and finally settled on kappa value of 5 which now reports the hot pixel percentage of 3.4%
I'm not sure if upping the kappa value that high is a problem. I guess the next time I process some lights I will look at the difference between using a BPM and not with that value.
Sorry, yes in this case it'll be the other way around. Lowering it will make it more sensitive to a signal. Yes, the only way to really verify will be to zoom in on a single sub, at an area with bad pixels, and see if it corrects them properly by switching to "l-calibrated". If you still see more bad pixels, the sensor will indeed have them and you need to lower the kappa on the BPM creation.
I've left it the BPM kappa level at 3 and I'm still around 8.5% bad pixels when creating the BPM using only darks, no flats. My BPM darks were 600s for my ASI294MC Pro. I can up to kappa level 5, but honestly I can't tell a lot of difference between the two levels switching from calibrated to l-calbrated on my lights.
I guess I'll leave it at kappa 3 even though it seems to report a high level of bad pixels which worries me.
Well that could simply be the case with your sensor. Hardware can have faults sometimes, I would watch it closely and every few months make a new BPM just to see if it stays the same. If so and you see no problems and it corrects properly, it'll be fine.
