19 June 2021: Our upload server https://upload.astropixelprocessor.com/ has been migrated successfully to our new office with higher upload and download speeds (nearly 10MByte/sec up/down ) ! We now have 1 general upload user called: upload with password: upload. The users upload1 - upload5 have been disabled.
31 May 2021: APP 1.083-beta2 has been released ! APP 1.083 stable will follow soon afterwards. It includes a completely new Star Reducer Tool, New File Saver Module, Improved Comet registration and much more, check the release notes here!
Difference between Astro Pixel Processor Normalization/LNC and PixInsight Normalize Scale Gradient Script?
Dear Mabula and Vincent,
Recently I've started to look into the PixInsight Normalize Scale Gradient Script. Really want to understand the difference, advantages, and disadvantages of both methods provided by APP and PixInsight.
PixInsight Normalize Scale Gradient Script looks quite promising to deal with some common hard issues - straight light, high clouds, etc. As far as there are light frames that could be used as the reference frame, the NSG script could do some really nice calculations and normalize the rest of the light frames collected in different situations (moon phase, clouds, stray lights for certain angles and disappeared after the flip).
Is APP using the same algorithm to calculate the normalization data? I would really appreciate it if you could elaborate more details about how the normalization (including LNC) is done in APP?
@mxcoppell I would really like to know about this as well.
Would also like to know when BETA3 will be available so drizzle issue is fixed, taking a long time.
I'll pass this one on to Mabula as he is the algorithm wizard. 😉
So, not sure why when blinking through my APP saved normalized light files my backgrounds brightness is not all the same. Does this mean when I save my normalized lights they are not actually normalized correctly??? This is Ha data from three nights of imaging. Here are two videos using ASI Studio to blink through my FIT files on both APP saved normalized lights and also normalized lights created in PI using the Normalize Scale Gradient Script. I was trying to upload videos but they kept getting blocked so here are the download links.
Blink through PI NSG Light Frames: Note: Background seems to match on all frames.
Blink through APP Normalized Light Frames: Note: Irregular background brightness
Thanks for sharing, Bill. In my eyes, the NSG is doing a fantastic job.
So I think I know why APP normalize lights was not consistent in my video. I found out that if I have multiple sessions with each session I had saved normalized lights, when combining the all normalized lights to make a final image, I loaded the lights and repeated the normalization process again so all sessions matched, which did not work out so well. It should have but for some reason it didn't, so I don't know if something is being left out. So this time I went back and cleared my data, then loaded calibrated lights from all sessions and then went through steps 3-6. I saved all my normalized lights and all had the same background brightness, so that's good!
I really hope APP could produce something like PI's NGS. It would be nice to also scale the lights the same. On my normalized light data, all the back grounds values are near the same but the nebulosity brightness (depending on my seeing) was fluctuating up and down, which I would fear the higher values (stronger signal) would get clipped in rejection process. It would be nice if APP could stretch lesser data sets to closure of that of the good sets which I believe NGS helps do this a little better and to also remove gradients in an automated process. The down side of NGS is that it doesn't seem to work with data needing drizzle, and I need that option.
Another thing I noticed through all this testing, if I run star analysis again on normalized lights, the image that was originally a reference frame was now scored lower. So should the normalize process be run a second time after a second star analysis run so a better reference frame is used??