NSG vs Local Normalization in PixInsight!
Vložit
- čas přidán 5. 09. 2024
- In this video I take a look at NSG vs Local Normalization in PixInsight. Which is better or is one easier to use? What do I use?
#PixInsight #Astrophotography
Help support the channel! Buying astro gear use these links below and I'll receive a small commission at no cost to you:
OPT (USA): bit.ly/37ZCnzv
High Point Scientific (USA): bit.ly/3CMPKEy
Ontario Telescope (Canada): bit.ly/3cncCjE
All-Star Telescope (Canada) bit.ly/3PZFyxI
New to Pixinsight image processing? Check out my PixInsight for Beginners Tutorial. 3-hour complete lesson from start to finish! More details here: bit.ly/3zAMqsF
Join me on:
Facebook: / visibledarkastro
Twitter: / visibledark_
TikTok: / visibledark
Thanks for watching! Clear skies!
I've been using NSG but after seeing your results, I'm willing to save the steps and time and use Local Normilization.
Thanks for answering this question, Shawn. Just using LN in WBPP now out of simplicity. NSG is a time consuming process, and hardly is worth the time considering ABE or DBE will pull out most gradients later anyway.
I've been using the local normalization since it became part of the WBPP script. I had tried NSG a few times, but the either/or decision of drizzle vs. NSG made NSG a non-starter for me. I really really like the drizzled output/improved star shapes (which is quite noticeable when using Dynamic Star Alignment).
I haven't tried local normalization. But I will. Thanks!
Hey Shawn, thanks for the great and very timely video. I think many people are struggling with the exact same question you raise here. One additional question though: Did your light frames vary a lot between them with e.g. clouds drifting through or strong differences in gradients from light pollution etc.? As far as I understand both processes (NSG and PI's Local Normalisation) are meant to address that specific situation mostly, meaning at you won't notice much differences between them or even between any of them and doing no local normalisation when your light frames are all reasonably homogeneous.
Shawn i totally agree with you, i have the paid version of NGS and compare the two and the one in WBPP is as good as NGS and less time consuming !
Hi Shawn, I've done some of those same tests and came to similar conclusions. NSG was clearly superior to LN for PixInsight versions prior to 1.8.9, but the revamped tools we got in 1.8.9 really closed the gap. I've found it easier to evaluate this on broadband (galaxy) pictures tough. I don't bother with NSG anymore now.
Very good video Shawn. Thank you
I’ve been waiting for this video. Thanks for making it. I use local normalization because it’s part of weighted batch processing. Call me lazy or whatever, but weighted batch processing is central to my workflow. I like the results using local normalization, but full disclosure, I’ve never used normalize scale gradient. Are you saying now that drizzle is completely integrated into we did batch processing? Does weighted batch processing spit out a drizzled image? I thought there needed to be another process for a completed drizzled image outside of weighted batch processing.
You still have to do the drizzle integration afterwards to produce a drizzle master. Local norm as part of WBPP though I find very useful. Easy to do with WBPP being central to my workflow as well.
Thanks for this video. One question: will NSG work better where I have some subs from the city and some subs from the dark sky site? I thought that is where NSG shines.
I must live in a great place (Bortle 5) as I rarely have any noticeable gradient where I live perhaps due to the fact that I rarely image over the south where our city lights are the most intense. I just never saw a big enough improvement in NSG to warrant all that extra work and CPU time for something that I couldn't distinguish.
I've stopped using LN last year because the results were not good - it sometimes produced a weird grid pattern in the background - particularly with a DSLR. I tried NSG several times but could not see much if any benefit for the long time it took to process, and then you can't even drizzle with it (that is a showstopper for me because I use an OSC camera with a small sensor). So, I might try LN again from WBPP as you described to see if it makes a positive difference. I always use WBPP with cosmetic correction to calibrate and debayer my OSC frames and get consistently good results, so checking the LN box is an easy step to add.
Hi Shaun, first thanks a lot for the time you spend creating this content. I have 2 questions : first, what was the amount of gradient on the original picture before you processed them and more importantly was it the same gradient on all the subs ? while i would tend to agree in case of identical simple gradient, i find NSG rather useful when for instance i have a light pollution source causing a simple gradient on a single frame but transformed in a quite complex gradient while integrating because the angle on each subs changed over time . Second question: a friend of mine suggested than better than NSG was actually to run ABE on the subs after cosmetic correction. He is using container to automate the action on all the subs. Did you ever try that and if so what you think ? Again, i am not suggesting that for all the images, only the ones with a complex gradient induced by aggregating simple gradient. I do realize this is not aligned with making less steps but it can save a lot of time later messing with background extraction. Thx
I’ve been using NSG With drizzle. I had some poor masters with local norm (although that was doing it manually and not WBPP)
Also I figured With multi night data I should local norm all the data at one ? That makes WBPP not possible unless there is some way to do multi night without turning 3 times widdershins at midnight under the golden moon while smoking half a dead mouse
Hey that's a Great Comparison of the 2 Shawn.. I Did find in the presets ? It also works in both the Maximum setting and Faster with Good quality. Including Drizzle Data with Both.. Ive played with it a few times And am Liking using Local with Drizzle a LOT ! . The Masters have been superb so far to be honest.
Hey Dylan. Thanks for the comment and insight!
@@VisibledarkAstro Welcome My Good man! It was a Great comparison Honestly! And the Timing was perfect.. I was Just Reading last night, the NormalizeScaleGradient (NSG) scrip and page had Moved . This Video convinced me to Continue the direction i was Headed with the Presets in WBPP already. So Thank you for that affirmation.
Great video , I have a question, local normalization will be bad for faith details, ? Ex: if I use Ha to get the background Ha from a Galaxy and is very faith signal, local normalization will not help , will simple normalize the back ground and the background detail will be lost , ?
Hi Shawn, great video and thanks for all the work you do. Without which a lot of us would not be able to make much progress. I have been witn Pixisight for a which but only just started to try and use NSG. I started to use it on some old data, 2 different cameras obviously with different pixel sizes but on the same scope. Any advice on how to use NSG and add all the output subs from the cameras into the same image. I simply ran NSG for each camera's batch of subs separately, and used PixMath to integrate them. However, I'm not sure this gives a correct answer. Feedback would be appreciated. V many thanks
Hi Alec - I have to say I don't know off hand. I think you would have to run the different batches of subs seperately. I suppose you could throw them all in together into NSG but you would need to align all of them first. The easier way to do it is to use WBPP and local normalization. You can add everything into WBPP from all nights, different scopes, and run it. Local normalization will be applied and no need for extra steps as NSG would require. Sorry I can't be of better help.
@@VisibledarkAstro thanks Shawn, response appreciated. Keep up the excellent work. Take care. Alec
Thank you for the review. This is a question, I have been asking myself for some while.
Can you reject artefacts, like dust donats with LocalNormalisation if you have a data set in the same channel, e.g. from a nother night, without those artefacts?
Thank you und clear skies,
Christian
Hey Shawn! I’m curious to know if you used the default settings or if you manually configured the gradient correction in the nsg script?
WBPP fails more than it works for me. So haven’t really tried either.
Hmmmm.... sometime it can fail of noise is an issue. Even of noise shouldn't be an issue. Running the light frames through cosmetic correction for hot/cold pixels, can sometime help correct this..
@@VisibledarkAstro the weird thing is I can stack the same data in asiair studio then transfer to PixInsight with no issues. That’s been my workflow recently, though I keep trying WBPP.
Hi Shawn, Great video but I do have a question,,, How do I update to the new version of wbps. I have tried to find a way and can't get it .. My pix just won't update that script, any help would be greatly appreciated Thanks.
Ha, please ignore the above it was an" ID-10-T" user error. LOL