Fantastic video. It is interesting to see different strategies while processing images. The way you handled color noise was amazing. I never would’ve thought of it.
Really nice tutorial! It makes me realize how much I still have to learn in pixinsight. Thanks for all the work you’re putting in to make pixinsight more accessible!
Really good Frank some steps in here I never knew were possible and others I never even knew existed. Will follow on again for later image processing ❤
Great video, thank you. I have learnt lots just by watching. I have been using your tools for a while now and they are fantastic. I only shoot OSC, so some of these steps don’t apply to me, but the process flow you have done and the tricks and tips shown I can certainly use.
Thank you!!! Amazing!! Please create an RGB image staring point and do a full processing video folding in your tools into a workflow that will help your many OSC camera members. Thanks so much Frank!!
Maybe I need to do a beginners course on masks and how to use them. In the mean time Jurgens Toolbox Selective Color script is a big help. I have a whole video on Selective Color too ua-cam.com/video/o7rVbNz_Dgg/v-deo.htmlsi=vkz7Vbm6wIUAsKrt
There is a great script to extract narrowband data from OSC and duo band NB filters. DBXtract pixinsight.com/forum/index.php?threads/new-script-dbxtract.23344/ if that is what you have this may help!
so cool! I have never seen that process container thing - I have to learn how to do that! quick question - why don't you start by combining the 3 channels in SHO and then doing all the steps? and auto linear fit... just curious
Becuase for narrowband you get way more control by strerching individually. Notice i had different target medians for each channel. Autolinear fit just does an unlinked stretch between them. I wanted each channelstretched differently. Even if I was combining them, I would still use Statistical Stretch in unlinked mode instead of autolinearfit. That would be redudant. Autolinear fit just isnt needed as much as people think!
Awesome video. Thank you once again!A few points: Is there a reason you didn’t choose the darkest part of the HA as your target median for stretching? You picked a bland grey area, but not the darkest part. Do you think it would be valuable to add background model to the preview for ADBE? I’m excited to try color masks…but I think I’ll have an easier time doing them in photoshop. A very powerful concept You’re killin’ this UA-cam thing!
I chose a dark grey region and not the darkest becuase on the OIII image i wanted the median value to hit around that level. If the median hits that dark but not darkest area that means there are still darker regions in the oiii image. If you match to the darkest region the oiii image wouldve had even darker spots. Adding a background model in the preview would be time consuming to run, it has to do the full subtraction in the background with copies of everything open temporarily. If you have large files that can start eating up ram quickly. Maybe if a lot of people want something like that I can look into how to implement it. Some items are probably much easier in photoshop. I dont own it and use GIMP for some items so I stick with the bulk of my processing in PI.
Working on analyzing FBPP now too. It is harder than I thought. FBPP produces such an apples to oranges comparison to WBPP it isnt even funny. I can think of one reason right now to use FBPP, as a preview of the data...
Excellent tutorial! What are your thoughts on Bill Blanchan's scripts 'Screen stars' (as opposed to Combined Images/PixelMath which you feature in the video) and 'StarReduction' for a finishing step?
Unnecessary now. His screen stars is pixel math with no preview, use the script, you can even adjust how much amount applies the second image. And StarStretch and NBtoRGBstars you can really fine tune the stretch on them.
I'm kinda wondering how come you didn't use the Dark Structure Enhance tool to give it more contrast? I find it to work pretty well. I just change one default setting (40) down as needed. Great video!!! I love using your tools 😊
I use that a lot too! Usually on galaxies and you are right you have to turn down the amount. Guess I missed featuring that script in this video. I do have it on some other tutorials I did.
Excellent Tutorials Frank, both narrowband and OSC! I'll be referring back to this one many times to hear the Master speak. Now for the first of many stupid questions: If I ever knew I can't remember how to get the levels readout on the cursor, tried many alt, shift, and ctrl combos but it still eludes me. Anyone willing to drop a hint?
Edit -> ReadoutOptions then make sure the Show readout preview is Checked. Then hit the blue circle to save it. Then you can hit the X to exit the Option window
Hi Frank just a quick question It would be really handy to have those image Identifiers you have to change the ha OII SII.How do you make them....sorry for the silly question
I dont have a Patreon. Only thing I have right now is youtube memberships. I am kicking around the idea of starting a patreon but unsure if I should or not.
They are a must for simplifying a bunch of pre processing! I have a video on them, but since a couple PI updates they have improved them. A blank ProcessContainer is found under All Processes and the checkbox to overwrite files is fixed ua-cam.com/video/8xmsWGyye28/v-deo.html
The colors were all wrong when you used Narrowband Normalization for SHO. I think you just did too much with the data prior to NN. When I have processed SHO data all I have done is channel combination, background neutralization and an unlinked stretch. You had to spend a lot of time rescuing the image after running NN. There is a workflow that Bill Blanshan recommends prior to NN to insure that the results in NN are correct.
Unlinked stretch is the same as stretching them seperately. I just think NN gives suboar results for SHO data. I have seen it on HOO as well and I am not a fan. That is a personal taste though
Wow your skills are amazing. Thank you for showing all these brilliant techniques. I have learned so much watching this.
I learned a lot watching this, thanks.
14:07 .. am I the only one seeing a reposed woman's head, side profile? 😃
This is the single-most useful narrow band processing tutorial I'vew seen...THANK YOU and joining the channel!
Thanks a bunch!!
Fantastic video. It is interesting to see different strategies while processing images. The way you handled color noise was amazing. I never would’ve thought of it.
Really nice tutorial! It makes me realize how much I still have to learn in pixinsight. Thanks for all the work you’re putting in to make pixinsight more accessible!
Love these kind of tutorials, as much as you look at diff kind of it there is always a new "touch" is some of them THX Frank !!!
That was excellent-even managed to help me salvage data I captured on the same target in 2022, when I was just getting started. Many thanks!
Glad it helped!
Really good Frank some steps in here I never knew were possible and others I never even knew existed. Will follow on again for later image processing ❤
Great video, thank you. I have learnt lots just by watching. I have been using your tools for a while now and they are fantastic. I only shoot OSC, so some of these steps don’t apply to me, but the process flow you have done and the tricks and tips shown I can certainly use.
Excellent video, and it has given me a plethora of new tools and ideas to try out. Thank you!
Damm! Fantastic!! Thank you so much for your fantastic scripts and tutorials!! You help me a lot to improve my astrophoto
Thank you!!! Amazing!! Please create an RGB image staring point and do a full processing video folding in your tools into a workflow that will help your many OSC camera members. Thanks so much Frank!!
I could do something like that. Do you have a OSC master you would like me to process on a video? You can email frank@setiastro.com
@@setiv2 absolutely....I will look through my files and send one via google drive. Thanks!!
Loved it! I started using PI only 10 days ago so using masks feels a little hard to handle right now
Maybe I need to do a beginners course on masks and how to use them. In the mean time Jurgens Toolbox Selective Color script is a big help. I have a whole video on Selective Color too
ua-cam.com/video/o7rVbNz_Dgg/v-deo.htmlsi=vkz7Vbm6wIUAsKrt
@@setiv2 Thank you ! As the new UA-cam star in astrophotography I didn't expect you'd have time to answer, very much appreciated
Really enjoyed this clip!
Thanks for posting
Cheers Simon
Fantastic video. Learned a few techniques. I really need to get a mono camera.
There is a great script to extract narrowband data from OSC and duo band NB filters. DBXtract pixinsight.com/forum/index.php?threads/new-script-dbxtract.23344/ if that is what you have this may help!
You always give very good ideas, thank you!
Very good tutorial Franklin! 👍Thank you!
Great tutorial using your latest tools. I plan to release a similar video soon using your tools but this time for OSC images.
Great Frank this is very useful thanks
Fantastic thankyou will join your Pateron
Very useful, thanks for sharing and for all your tools :)
really interesting I downloaded your scripts and they are very interesting. Compliments
so cool! I have never seen that process container thing - I have to learn how to do that! quick question - why don't you start by combining the 3 channels in SHO and then doing all the steps? and auto linear fit... just curious
Becuase for narrowband you get way more control by strerching individually. Notice i had different target medians for each channel. Autolinear fit just does an unlinked stretch between them. I wanted each channelstretched differently. Even if I was combining them, I would still use Statistical Stretch in unlinked mode instead of autolinearfit. That would be redudant. Autolinear fit just isnt needed as much as people think!
Awesome video. Thank you once again!A few points:
Is there a reason you didn’t choose the darkest part of the HA as your target median for stretching? You picked a bland grey area, but not the darkest part.
Do you think it would be valuable to add background model to the preview for ADBE?
I’m excited to try color masks…but I think I’ll have an easier time doing them in photoshop. A very powerful concept
You’re killin’ this UA-cam thing!
I chose a dark grey region and not the darkest becuase on the OIII image i wanted the median value to hit around that level. If the median hits that dark but not darkest area that means there are still darker regions in the oiii image. If you match to the darkest region the oiii image wouldve had even darker spots.
Adding a background model in the preview would be time consuming to run, it has to do the full subtraction in the background with copies of everything open temporarily. If you have large files that can start eating up ram quickly. Maybe if a lot of people want something like that I can look into how to implement it.
Some items are probably much easier in photoshop. I dont own it and use GIMP for some items so I stick with the bulk of my processing in PI.
@@setiv2 that all makes perfect sense. Thank you!
Working on analyzing FBPP now too. It is harder than I thought. FBPP produces such an apples to oranges comparison to WBPP it isnt even funny. I can think of one reason right now to use FBPP, as a preview of the data...
Excellent tutorial! What are your thoughts on Bill Blanchan's scripts 'Screen stars' (as opposed to Combined Images/PixelMath which you feature in the video) and 'StarReduction' for a finishing step?
Unnecessary now. His screen stars is pixel math with no preview, use the script, you can even adjust how much amount applies the second image. And StarStretch and NBtoRGBstars you can really fine tune the stretch on them.
I'm kinda wondering how come you didn't use the Dark Structure Enhance tool to give it more contrast? I find it to work pretty well. I just change one default setting (40) down as needed. Great video!!! I love using your tools 😊
I use that a lot too! Usually on galaxies and you are right you have to turn down the amount. Guess I missed featuring that script in this video. I do have it on some other tutorials I did.
Nice work!
Excellent Tutorials Frank, both narrowband and OSC! I'll be referring back to this one many times to hear the Master speak.
Now for the first of many stupid questions: If I ever knew I can't remember how to get the levels readout on the cursor, tried many alt, shift, and ctrl combos but it still eludes me.
Anyone willing to drop a hint?
Edit -> ReadoutOptions then make sure the Show readout preview is Checked. Then hit the blue circle to save it. Then you can hit the X to exit the Option window
@@setiv2 El Perfecto 👍 Thank you Sir!
Great video!
Commented. Liked. Subscribed.
Hi Frank just a quick question It would be really handy to have those image Identifiers you have to change the ha OII SII.How do you make them....sorry for the silly question
You can get my processing icons here: drive.google.com/file/d/150KYImZNUEuGS5XyafH2YoM21Gbexi-y/view?usp=drive_link
@@setiv2 fantastic thank you.Will join Patreon
I dont have a Patreon. Only thing I have right now is youtube memberships. I am kicking around the idea of starting a patreon but unsure if I should or not.
Were the photos pre-processed (histogram) at the very beginning? Are these raw frames after Stacking?
Raw after stacking
Interesting.
How I am just learning about Process and Image containers?!!
They are a must for simplifying a bunch of pre processing! I have a video on them, but since a couple PI updates they have improved them. A blank ProcessContainer is found under All Processes and the checkbox to overwrite files is fixed
ua-cam.com/video/8xmsWGyye28/v-deo.html
The colors were all wrong when you used Narrowband Normalization for SHO. I think you just did too much with the data prior to NN. When I have processed SHO data all I have done is channel combination, background neutralization and an unlinked stretch. You had to spend a lot of time rescuing the image after running NN. There is a workflow that Bill Blanshan recommends prior to NN to insure that the results in NN are correct.
Unlinked stretch is the same as stretching them seperately. I just think NN gives suboar results for SHO data. I have seen it on HOO as well and I am not a fan. That is a personal taste though
That’s because you are not using NN correctly. I get great results with NN for HOO and SHO.