BlurXTerminator with Russ Croman (Interview, Use, Workflow and more)

Free 30-day trial (or purchase) of BlurXTerminator:
www.rc-astro.com/resources/Bl...
Here is the first interview Russ has given since he released BlurXTerminator to the world in late December 2022. This is a wide ranging interview that covers many areas and addresses many questions through Russ's explanations and demonstrations.
As mentioned in the video a comprehensive explanation of the architecture of Russ's neural network can be found in this AIC video he published on his own KZbin channel at: • AI for Astrophotograph...
A shorter version of this video can be found here:
• BlurXTerminator with R...
Use the Chapters below to find topics of interest:
00:00 Introduction (Interview and Explanation)
08:10 Tour of the BXT process by Russ
09:22 Explanation of Automatic/Manual PSF for NonStellar features
15:00 Demonstration of variable sharpening (Automatic PSF)
16:17 What is oversharpening? (Fine scale deconvolution)
20:05 Behavior of the PSF Diameter Slider
20:45 How to determine best PSF value
22:20 Major Updates to AI version 2
25:05 Luminance Only Mode and Aberration Corrections
26:51 Chromatic Aberration Correct Only (demonstration)
29:18 Manual PSF uses circular/symmetric PSF (demonstration)
29:50 Luminance Only Mode Demonstration
33:25 Correct First (demonstration)
35:21 Best Workflow with BXT
41:57 RGB Considerations
42:42 BXT on individual color channels?
43:47 BXT For matching PSFs (repair of images)
44:54 LRGB considerations
46:10 Narrowband Images (SHO)
51:00 Demonstrations in PixInsight
51:40 Baffled at BXT Popularity?? (SNR example)
55:11 Limitation of Deconvolution
56:24 Comet example (using BXT, SXT, NXT...)
57:35 Batch mode of SXT mentioned
58:57 M88 example (LRGB)
59:41 Showing improvements of the different AI versions (and Luminance Mode)
1:02:05 Ill matched PSFs for LRGB (example)
1:04:13 SH2-104 example
1:05:00 BXT only sharpens features that exist in image
1:09:02 BXT is not "content aware" filling features
1:11:45 Overfitting of Neural Net?
1:18:07 Results (vdB 1)
1:19:46 Results (vdB 4)
1:22:09 Final words Wrapup
More at:
AdamBlockStudios.com
See the Reviews:
www.adamblockstudios.com/cate...
Those just starting out in PixInsight will enjoy FastTrack Training!
(Updated to include SPCC!!!)
www.adamblockstudios.com/cate...
(see the video • Introducing FastTrack ... )
#PixInsight #BlurXTerminator #NoiseXTerminator #StarXTerminator #Deconvolution #neuralnetworks #vdb1 #vdb4 #sh2106 #AdamBlockStudios

[@EthansDeepField]

Thanks Adam and Russ, I'm grateful that the community have guys like you to keep things moving, your videos are always helpful. The thing I love the most in this video is the moment Russ tell you there is no immediate next project, your smile slowly disappeared, and reappeared when he tells you he is thinking about AI gradient removal tool, I'm on the same emotional rollercoaster with you!

[@AdamBlock]

haha!

[@DBFIU]

Adam and Russ killing the game!

[@Xanthus723]

That was a very good watch thanks to the both of you for what you contribute to this community!

[@astrofotoperu]

I have Blur Xterminator and Noise Xterminator. Both are my favorite Pixinsight modules. They are essencial in my processing steps. Thanks Russ! I'm already anxious to try your next project: Gradient Xterminator! Also, thanks for the interview Adam! Great info!

[@FrancoGrimoldi]

This is pure gold. Thanks!!!

[@SassePhoto]

Wow - this is a paradigm shift in astrophotography. Excellent interview, many thanks!

[@AdamBlock]

Thanks for watching!

[@NeilCorke]

Thank you for posting this Adam. Russ has always been helpful and I have all his plugins, so bought Blur XTerminator straight away when this was released.

[@dow4hurst]

The Dwarf Lab's Dwarf II telescope has reawakened my interest in astrophotography. I've taken the leap into Adam's PixInsight tutorials, along with trial licenses on Russ Croman's 'XTerminator series of software. I will likely pull out the Celestron 8" EdgeHD as a result of this experience! Thank you for this wonderful video explaining the point spread function based deconvolution features and uses in BlurXTerminator. (And, when not to use it!)

[@jonathanpearceff]

Great informative interview. Something to watch a couple of times to fully understand how to get the best from this amazing tool.

[@astrocrypto8438]

Excellent video. Thanks to both you, Adam, and also Russ. Amazing resources you both are.

[@AdamBlock]

Thank you for watchign!

[@alancreutz9561]

Really good discussion of really good stuff. Thank you both!

[@peruzatto]

Excellent interview. Thank you both.

[@AdamBlock]

Thanks!

[@berndlimburg134]

Fantastic, thanks Russ and Adam for this great and so informative interview! 😀👍

[@AdamBlock]

Thank you!

[@MrPedalpaddle]

Excellent interview! Thanks for the chapter markings, work flow discussion, & SHO discussions

[@AdamBlock]

Great! I will be posting another bonus section from this interview... there is still more material if you can believe that.

[@scottjohnstone2001]

Excellent talk. Touched on every single topic I would have asked myself. Thank You.

[@AdamBlock]

Excellent!

[@blanejnasveschuk6351]

Super interview and video for BXT! Excellent review of design, functionality, use of BlurXTerminator and scrutiny on results, dispelling unfounded rumors! Can’t wait to use! TY

[@mauriciomorenorivera]

Love this

[@ysl109812]

Such a useful video. It is tremendously helpful. Thank you both!

[@bnkohrn]

Very insightful video. Saving this for reference was an immediate reaction. 👏🏻 EDIT: wowzers, that vdB4 image is stunning!

[@AdamBlock]

Oooh... thank you Nick!!

[@billblanshan3021]

Great job guys!

[@AdamBlock]

Thanks BIll.

[@terrizittritsch745]

Great interview, thanks Adam. I have to admit, the first time I used BlurXTerminator on some of my data I was surprised by how much it did, with so little effort. I guess I shouldn’t have been surprised as I have used GradientXterminator for quite a long time in photoshop. Great to see you two Goliath’s in this space team up.

[@davewilton6021]

Excellent interview! The part I found most valuable was the discussion of the order of the workflow. But the whole thing was great.

[@AdamBlock]

Thank you so much for watching... quite a bit to absorb!

[@luigimorrone9401]

Adam many thanks for your great presentation . Russ Croman tools are Amazing!!👍😉

[@AdamBlock]

Thank you!

[@ferashaki7528]

Great video!

[@AdamBlock]

Thank you!

[@michaels4668]

Wow,that was a fantastic, informative, and classroom like interview. I still a old novice learning processing .

[@deep_space_dave]

Thank you Adam for putting this together. This video really shows that a tool can only be good when used properly. Also tell Russ thanks to him I had an excuse to buy a new PC workstation with a RTX A5000 😀

[@harrysastroshed]

Excellent information thanks for putting this out - the first tool I ever bought was grad x many years ago - Lets hope russ has many many ideas🤩

[@kentblackburn412]

I’d be interested in Ross’ thoughts on drizzle in combination with BlurXTerminator.

[@johannesrussek4672]

Thanks for a great video! One thing I'd love to understand a little better is how you matched the PSF between L and RGB @1:02:05 - what process did you go through to find the similar settings?

[@AdamBlock]

For LRGB the match does not need to be exact..but there isn't a precise numerical method. Pretty much look at the RGB results (perhaps measure the FWHM if you want). By eye is good enough.

[@matthiassteinmetz6394]

Excellent video, Adam and Russ, thanks a lot. How would your workflow be for HaRGB, if I may ask (e.g. Ha regions of galaxies). Usually one adds a continuum subtracted Ha to R, followed by SPCC. BXT before would violate no SPCC after BXT, while BXT after would violate no BXT with mixed PSFs (here R and Ha). Any suggestion?

[@AdamBlock]

I think you would apply BXT to each of Ha an R before subtracting anything. You would need to be careful of course... probably using the same manually set PSF setting. This is a guess.

[@blueh2o575]

Great Video Adam, super helpful. It seems the community has much to offer in learning the best methods for BXT use. In the spirit of that, I'd like to post two questions: 1. Regarding your question at 49:00s about a recommended workflow for RGB stars that are replaced into NB images, I wonder why Russ recommended to remove the stars first (and use the average FWHM determined prior from the NB stars). It seems to me that using BXT first before removing the NB stars would allow one to use the Automatic PSF (and it's benefit then of correcting non-stellar parts of the image) and only then to remove/replace the NB stars with RGB stars. 2. Regarding multiple data sets taken over multiple nights and using multiple exposure times....Since the NN is taught to determine the PSF created by coma, atmospheric distortions, tracking errors, etc, is it really a good idea to apply BXT only to the final integrated image? Since the PSF is likely unique for each of these data sets, I would expect it to be better to apply BXT to each set of integrations which are grouped by imaging session and even possibly those grouped by exposure time. One issue with this method that I worry about is if the mass centers of the stars might shift due to the the BXT corrections, preventing a good image integration of these data sets. Thanks so much for any inputs on these thoughts.

[@AdamBlock]

I think you are right concerning #1. However, a cautionary point is that many NB images are actually star poor sometimes. If so, then measuring the stars and using that value for the manual PSF is good. At the end of the day... I do not think it is too critical. With regards to #2 my intuition is that you should really stick to the full integrated image in general. Deconvolution relies on S/N first and foremost. In addition, if the star aberrations and seeing kinds of average out into a circular-like PSF... then no problem. If however, the aberration such as coma or astigmatism remain in the integrated image (much more likely)- then BXT can still take care of things. I do not think there will be a benefit. You should (could) always do the test/comparison and tell me I am wrong. (but I really do not think so here)

[@blueh2o575]

@Adam Block I appreciate those insights, very helpful. It will be interesting to see the best known methods as the community experience with BXT grows.

[@garythompson3869]

Great video Guys. Understanding the why? gives you the what?

[@malcolqwe2]

Adam, I typically shoot 1x bin LUM, 2x bin RGB (old ccd). I BXT the LUM only, then delinearize and combine LRGB. Are you saying in this video that I ought to BXT both the Lum AND RGB before making the LRGB. thanks. It sounds like there's a chromatic aberration routine BXT can do to the RGB that I'm missing out on

[@AdamBlock]

Yes, you should run BXT on the RGB for two reasons. First is to better match the PSFs. You want your stars in the RGB to be about the same size as in the Lum. Second, as you mention, BXT comes with additional benefits of aberration correction.

[@malcolqwe2]

@@AdamBlock thank you :)

[@cemoguz2786]

This is funny Adam Block is comparing his resulution of telescope to a real big goverment founded telescope like I compare mine to a 17 inc telescope. :D