Pixinsight noise reduction tutorial 2022 (my new technique and trick!)
ฝัง
- เผยแพร่เมื่อ 6 ก.ย. 2024
- In this video I'll show you my current technique for PixInsight noise reduction using Multiscale Median Transform plus another little trick.
Download the MMTNR process icon here: amzn.to/3CSS7V3
Help support the channel! Buying astro gear use these links below and I'll receive a small commission at no cost to you:
OPT bit.ly/3xjFo9t
High Point Scientific: bit.ly/3ueQtLo
KW Telescope (Canada): bit.ly/3gk8YWi
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
Handy! Telescope cable organizers here: etsy.me/3gi1VgR
Join me on:
Facebook: / visibledarkastro
Twitter: / visibledark_
TikTok: / visibledark
Thanks for watching! Clear skies!
![Warhammer 40k: Space Marine 2 | เพื่อจักรพรรดิ ! [ตอนเดียวจบ]](http://i.ytimg.com/vi/4M0ckYMA4-U/mqdefault.jpg)
In regard to the black spots, I think the Adaptive slider (under Amount) is for removing those, and it seems to work quite well too.
I’m going to have to give this a try. Lately, I’ve been using the EzDenoise script just before stretching. It gives fairly reliable results, but it is very very very slow.
I use EZ suite selectively. I've found the results often undesirable or not good for some processes. It has some useful purposes though. But I much prefer manually doing noise reduction.
Hi Shawn. Nice video! I find that when I am done processing with various denoising techniques, like the ones you have demonstrated, I still have the nagging problem of residual noise/artifacts in the sky background that don't seem to completely go away. This is particularly noticeable when I zoom way into the corners of my images. However, using Pixel Math, I think I have come up with a simple solution to equalize the sky background without impacting the main objects in my image. Here goes: After I am finished processing my image, I then examine my work to find the faintest possible detail that I want to protect (usually some dust lane or faint nebulosity). I then measure the brightness of that detail to establish a baseline. For example, if this faint detail appears to have a brightness (K) of 0.07, then anything dimmer than about 0.06 contains the pesky noise which I wish to eliminate. Now, using Pixel Math, I enter a single RGB/K expression as follows to equalize the background: iif($T
That's really interesting! I tried it. You do have to be careful with it though as you wrote. Nice result though. Thanks for sharing this!
Wow. This is by far the best results I have had for noise reduction. Captured with a Canon Eos R on a ~90 degree night in Texas. This stomped out the majority of the noise. Thank you for taking the time to share!
Thanks Shawn! Definitely will check
Thank you once again for a very useful instructional noise reduction technique. You are my go to now for Astrophotography image editing. Your explanations for using PixInsight are simple, straightforward and enjoyable to watch. Keep up the good work.
Wow, thanks!
seconded. Shawn is too gracious methinks.
Once again you have knocked it out of the park. I just messed around with this on my Christmas Tree cluster image and wow. I buckled down back in June and made myself use Pixinsight and I have really made progress. I owe much of that to you. Thanks!
Excellent tutorial as usual Shawn with logical flow and explanation. Your beginner tutorial helped fast-track my PI workflow and I continue to recommend it to others like myself, just getting started.
Hi Shawn.
I have just tried out your methodology.
Superb results, yielding buttery-smooth, noise free images!!
Loving it, please keep the great tips and videos going 👍👏👏!
Definitely going to try this later. Sharing now to Instagram. You’re tutorial has helped our family out tremendously. Thank you!
Your welcome! Thanks for sharing it. :)
I have tried several ways doing noise reduction an one is similar to what you demonstrated here. Sometimes just ACDNR on the stretched image works good enough, other times I have to spend a lot of time trying to find a way to remove the noise without cutting out too much details. Because some noise reduction tricks can truly butcher an image as well.
Other times I try to do it before stretching and sometimes I end up playing back/forth, as sharpening brings the noise right back lol!
Greatvideo. PixInsight is changing and improving in leaps and bounds. For example, weighted batch processing has changed dramatically recently.I would like to see you to do a processing update on weighted batch processing.
Your videos are extremely helpful & I always learn from them. I’ve even bought your latest pixinsite video & that really gave me a lot of new ways to do things. Thanks for sharing.
Wow! I was thinking EZDenoise was the holy grail of noise reduction without the hassle. Boy was I wrong (or I was really doing it wrong). This is WAY better for my relatively short integration times (a couple of hours with L-Extreme and OSC).
Thanks so much for the detailed explanation and the time you take to share.
I now do a MLT in linear mode, stretch, rest of my workflow, then a TGVdenoise and after that another MLT run which helps deal with the some of the residual TGV issues. My workflow has become even more complex lol. I have ran out of data to play with though (my imaging rig isn't ready for imaging yet, job/income issues have meant I have had to drop the hobby).
Shawn - Another outstanding tutorial. To date I have been doing almost all noise reduction in the Linear state. Thank You!
Thanks for the video, and marking it as 2022. Six and say it’s improving at such a fantastic rate, it’s hard to remember what’s new and what’s old. Would love you to do a video on the new weighted batch processing.
Awesome! I watch Adam's videos for the details and yours to see how to actually use PI ; )
Thanks for all the great content!!
Thanks for watching! Glad the vids are helpful. :)
nice explanation, perhaps you could explain the difference between using the channel combination and LRGB Combination tool??
Sure thing!
Awesome technique - thanks for the video
Very useful Shawn thank you!
Fantastic technique! I've been looking for a better noise reduction method and I liked it so much I ended up going back and using it on my latest 20 images. Thanks Shawn!
Thanks Shawn. I already use the Multiscale LT denoise methods from you previous tutorials and of course ACDNR and TVGdenoise, so this is an additional tool. I tried it out and it works well, and I have added it to my saved icons set. I don't know if I am right or not, but I usually pre-stretch and denoise the mono images before combing them to RGB. The theory being that if you start with good images then the final results will have less defects. I have found it works OK but I would be interested in your view, given you are a Pixinsight guru.
Bravo, Sean that is an excellent process
Interesting method, I'll have to give this a go
Super! Hope it works well for you.
thanks Shawn!
Great video :-) But can I ask a question? .... How did you go about making the perfect elliptical mask that put in place before you carried out ACDNR?
I use the GAME script for Pixinsight. It helps you create masks.
@@VisibledarkAstro Many thanks for your answer Shawn 🙂
Very interesting video! Thanks. But I don't know if it is TH-cam, my connection, or the video compression, it's very hard to actually see the noise in your images, because the video is very blocky most of the time.
great video, thanks!
Great tutorial as usual,,,, Thank you .
Glad you enjoyed it
Awesome. Thank you!
Glad you liked it!
The question I always want to ask you is that for MMT, how do you think the threshold for each layer is too much or too little?
Extracting the wavelet layers and inspecting them for noise is helpful. Can give you a good idea what layers have the most noise. And most often is layers 1 to 3. That being said, I just make threshold adjustments and check the result in a preview. It's often different for each image. I adjust them until I'm happy with the result.
@@VisibledarkAstro thanks Shawn!
Shawn
I shoot OSC 533 can I extract luminance and apply this technique to RGB only then pixel math luminance back in? Thanks!
Hi Shawn, sorry to trouble you. I have followed the link for the icon and for me I cant seem to download it ?
No problem. Not sure why its not working for you. You can email me shawn (at) Visibledark.ca and I can send it to you.
Shawn , question, I use a color cam the 294 zwo... Do you extract luminance from the image and then work on the rgb and then reincorporate it ? Not sure how to do the lum
I've a question concerning drizzle. I have Pix generate the drizzle data with each run of the WBPP, mainly because I find the removal of star pixelation attractive. The down side is that that the image file size explodes from around 50mb to just over 1gb with the corresponding impact on image processing times in the various tools within PixInsight (ZWO ASI2600MC-P). As I consider to consider going to a monochrome camera/NB imaging, that growth in file size for each filter channel becomes both daunting and a drag on storage resources. So I'm curious about your thoughts of drizzle vs. non drizzle and what impact drizzle has on you processing?
So complicated.. :-)
With PhotoShop plugins like e.g. Topaz Denoise AI this is done with one klick.. But I belive you belive your result is better ;-)
I absolutely hate it. The effect will be too strong, it will look like plastic
p̾r̾o̾m̾o̾s̾m̾ 😀