10.48
#...
·
·
2
likes
|
|---|
|
First off, this community has been so good to me, so I want to thank you all for your time and generosity in advance. I've been doing astrophotography for about 1 and 1/2 year first with a DLSR and for the past 2 months with a cooled mono camera with a 1600mm pro that I got second hand. I have been fascinated with space since I was a kid and this hobby combines my love of science and my passion for arts all in one... I feel deeply connected to this craft and deep down inside me I think I belong in this space I'm pretty good with computers and learning software. I use photoshop, Pixinsight and APP mostly. I'm looking for some Jedi advice on both a technical and creative stand point. I love taking and processing astro images and I'm trying to become as good as I can be. Recently I started to make more of an effort in framing things better , work on my colour palette and also write a little bit of a story with every picture I take. Without a story there is no picture. I'm trying to present data in a way that appeals to my eye and what feels right to me and also contribute to showcasing interesting details in the subject. Please be honest, if you notice something that you think is pretty average in my images, or anything that I could improve on, I would be eternally thankful for the feedback. Here are some of what I think are my best images to date. Thanks you so much for your time  NGC 3372 The Carina Nebula, Queen of all nebulas  IC 2944 The Running Chicken Nebula featuring the planetary nebula PK 294-0.1  Sh2-308 The Dolphin Nebula in HOO  Rosette Nebula in SHO  HorseHead and Flame Nebula in HSO  The Flying Frilled Neck Lizard Fire Dance - NGC 2264 in SHO |
4.52
#...
·
·
2
likes
|
|---|
|
Expressing Your critical or unflattering views on an astron is unwelcome. Sometimes it is very difficult to write anything substantially negative that will not be removed. For me on Your's image NGC2264z SH2-308 the noise is tooooooo big, and the stars are pull out from nebula.Stars are also too wild. IMO the best colors has Chicken, but the worst (unusual) Horse. Rosette have great cyans/blues levels. |
|
10.48
#...
·
·
1
like
|
|---|
Tayson: @Tayson Thank you for your honest opinion. I wasn't unaware than in the critic section people could not critic in a negative fashion without getting pulled.. especially when I'm asking for it I really appreciate your time and feedback and I can see from your images that you know what you are talking about  If you don't mind me asking for clarification on a couple of things. I understand about the noise, I probably would need more integration time. As for the stars being pull out of the nebula, do you mean the stars are too bloated? And by wild are you referring to colour/shape being wrong? Thank you again |
|
4.52
#...
·
·
2
likes
|
|---|
|
If You want smaller noise less stretch image , or more denoise image (not good idea), or the best take more data. The stars do not have nice bright core, shape are bloated like cirrus on sky. In a few parts I can see a tiny panda eye around stars. When You processing stars I thing better is to do less or to much. If You can't get nice colors of stars from NB channels maybe use BW version? Sharp black/white stars for me are much better than colorfull sharpless dots. |
|
10.48
#...
·
·
1
like
|
|---|
| @Tayson I get what you mean. Most definitely has to do with my processing for the most part. I'll keep what you said in mind and try to reprocess NGC2264 and see how I can improve on the stars and noise profile. Thank you |
|
4.52
#...
·
·
1
like
|
|---|
| Sometimes smaller resolution is also good way to dimmer photo imperfections. |
|
10.48
#...
·
·
1
like
|
|---|
@Tayson I have reprocessed this picture with your feedback in mind. I think it is a improvement in noise handling, colour and star shape. The Flying Frilled Neck Lizard Fire Dance - NGC 2264 in SHO |
7.53
#...
·
·
3
likes
|
|---|
|
It's hard to offer anything that is not subjective in terms of style which is a good thing. However I would say that if there is a general trend it is a lack of integration time, especially since you live in Bortle 7. You can see the effect the lack of integration time has in some of your pictures. You have great detail in the brighter areas with a better SNR, but its starts to fall apart once we move to the dimmer parts of the picture. This is mainly a symptom of the integration time as a lower SNR has more noise and less detail to work with so once you add de-noise on top of that things start to get blurry. You have some beautiful 3nm filters but you need to give them more integration time to let them do their thing. |
11.01
#...
·
·
1
like
|
|---|
Steeve Body: For me, personally, I prefer the original image. Overdoing noise correction does lead to a slightly milky feel to the image which doesn't feel right. Just a personal opinion. |
|
10.48
#...
·
|
|---|
It's hard to offer anything that is not subjective in terms of style which is a good thing. @SemiPro Thanks so much for your feedback. Yes indeed you are not the first one to say that and I'm sure collecting a lot more data would help. I think what is happening is that I get too excited to see what I collected, process it and then move on to another target.. it is a bit like a drug so to speak! I live in Melbourne and clear nights can be a rare thing... As for NGC 2262, I originally had 6h worth of data when I first processed it and decided to double it 12h to see if it would make any difference to the faint part of the nebula... and it made very very little difference I could see, so I started to doubt that more data would add anything and maybe I reached the point of diminishing returns... In saying that some of the best picture I see have between 20 and 80h worth of integration... so there must be merit in that but then I see guys like @EmuHead who consistently take 2h of data and make it look amazing! By the way Emuhead you were my inspiration for IC2944... so if you have some de-noising trick to share! please let us know! |
|
10.48
#...
·
|
|---|
Andy Wray:Steeve Body: Thanks for your comment, there is always a trade off isn't there!? I prefer the colours I got on the new one but yes I lost something in the faint background when de-noising the grainy texture... Probably overdid it... which is so easy to do! But I'm glad you chimed in on that so I have a better perception of what can be perceived as too much. |
9.48
#...
·
·
3
likes
|
|---|
|
For someone who has been doing astrophotography for less than two years, you have certainly gotten off to a good start. I find your images to be well-composed and aesthetically pleasing, and they will only get better as you gain experience. If there is one small thing in your processing on which to work, I would say that your brighter stars tend to look overexposed. In other words, they become flat circles after stretching to the point where the nebula looks good. You could look into various star size reduction routines, or learn how to work with starless images to create a final version with less highly stretched stars. You could also experiment with combining shorter exposures for the stars and longer exposures for the nebulosity to create a pseudo-HDR composite. The processing steps required to reduce star size are the key, even if your raw data are excellent. If you should look through Top Picks and IOTD, most of them have very nicely processed stars in terms of size, shape, and color, satisfying the technical criteria used in the review and judging process as set forth by Salvatore. Once in a while, something slips through the cracks with ugly stars, and peoples' feathers get ruffled quickly.  |
5.26
#...
·
|
|---|
Steeve Body:Tayson: |
|
10.48
#...
·
|
|---|
For someone who has been doing astrophotography for less than two years, you have certainly gotten off to a good start. I find your images to be well-composed and aesthetically pleasing, and they will only get better as you gain experience. Thank you so much for your kind words and feedback Rhedden. Yes I do seem to struggle with stars (and noise because of a lack of integration time) and this is something I really want to address so I can take my images to the next level. I do work on a starless image for most of the processing and then add the stars back at the end with Pixelmath. My narrowband stars are much tighter than when I shoot with RGB filters, but with RGB the colour are so much better for the stars... so I do try when I can to shoot 30s RGB stars and only transfer the colours to my narrowband stars since there RGB stars are more bloated than my 300s exposure narrowband shots... and that sort of work. I do this in photoshop by layering the RGB stars on top of my narrowband star mask and use the "colour" layer blending mode to only transfer the colour information but keep the narrowband star shape, then I bring that back in pixinsight and use pixelmath to combine the stars with the processed starless image. The main issue I'm having when doing this is that when using Star Xterminator or Starnet to extract the stars, some of the brightness information from the bigger stars remain in the nebulosity of the nebula itself. The starless image is clean with no stars but some of the bigger stars only regain their brightness information when combined back with the starless picture... because my starless image has been stretched and processed when I combine back with the stars this can create colour and brightness issues... maybe I should try 10s exposure instead and completely replace the narrowband stars with RGB stars? But I'm not sure if this is going to solve my problem, but definitely something to try like you suggested with an HDR composite just for the stars. I have included my integrated fits for NGC 2264 (since I believe this image is suffering in terms of bright star shape) here if maybe would wanted to check the data before I did any processing to it and see if the issue with my stars is in my raw data itself or in my processing... or both! https://drive.google.com/drive/folders/16f6phLC_DKOhv1RZ2jrqUneOCSOj8Sbi?usp=sharing The only thing done here is a slight cropped, background extraction as well as an EZ softstrech. I also included a no strech version just in case and an SHO combination from those 3 channels with no processing whatsover. Worth noting that for this image I did not use rgb stars at all, only the narrowband stars. Another thing I would like to mention is that I do try to use star reduction with EZ processing suite using morphological transformation or the Adam Block method, but I always notice artefacts so I have not used it much. The morphological transformation method leaves some little dark circles around the stars and the Adam block methods is better but still leaves a starnet artifacts around the stars... also tried the APP star reduction tool, and that really leaves tones of artifacts... I do remember getting better result in the past making the starmask myself and using morphological transformation based on a video I watched from VisibleDark on youtube... maybe I should try that again instead of EZ suite |
|
9.48
#...
·
·
1
like
|
|---|
|
I have all the same problems you do with star size, and we basically use the same workflow to transfer RGB color to the narrowband stars. I can't say that I have a quick fix for these problems. I will say that you should download Starnet++ v2 immediately, if you have not already, because it helps with the bright star removal problems. I opened your H-alpha data and took a quick look at the NO and EZ files. The EZ version already has the central star cooked to the point of clipping. The raw, unstretched NO version has a much smaller core in the central star. I applied a quick stretch to the NO version using Levels in GIMP just to brighten it, and here is the difference: EZ version (already stretched)  NO version (I stretched it with Levels)  One thing you could do is to open the NO version in Photoshop and try stretching it repeatedly with Levels, and then finishing it off with Curves. Keep your eye on the brightest star, and don't let it get bloated. Then send it off to starnet++ v2. After you have the starless image, finish stretching the nebula to your liking using Curves. That's exactly how I deal with stretching these days (coming from a GIMP user, rather than Photoshop). |
|
10.48
#...
·
·
1
like
|
|---|
I have all the same problems you do with star size, and we basically use the same workflow to transfer RGB color to the narrowband stars. I can't say that I have a quick fix for these problems. I will say that you should download Starnet++ v2 immediately, if you have not already, because it helps with the bright star removal problems. Wow that is a massive difference! Thanks for doing that! Manual stretch it is. I have always used the auto stretch function in Pixinsight not realizing that this was creating lots of problems. good to know that Im not the only one with these stars processing issues. I’m looking forward to giving this a shot in my next round of processing! I’m currently attempting (weather permits) to add 20h worth of data to my dolphin nebula to get a better SNR and probably going to start doing that with all my other targets before summer is up. you’ve been extremely helpful rhedden. Big shout out to you and thank you so much for your generosity with your knowledge and time. |
|
10.48
#...
·
|
|---|
@rhedden I have reprocessed this image following your advice and using Photoshop to do the stretching with levels and a bit of curve at the end. The stars are half the size of what they use to be and I have not use star reduction.  The Flying Frilled Neck Lizard Fire Dance - NGC 2264 in SHO Am I getting closer to acceptable stars now you think? |
|
9.48
#...
·
·
1
like
|
|---|
|
I'd be happy with that result on the Cone in terms of star size. It's really a question of personal tastes, though. Some of your other images already have nice stars; it's just a few of the brightest ones that got away (like in the Dolphin Head image). Some people reduce the star size too far, which is equally bad, in my view. Now that you have a handle on how to manipulate star size during stretching, do what you'd like with it, and don't be too sensitive to other peoples' preferences. The sizes are right when you feel they are right. P.S., don't look at any of my images taken with the C11 Edge. Some of them have stars that look more like a mosaic of planetary images.  |
|
10.48
#...
·
·
1
like
|
|---|
I'd be happy with that result on the Cone in terms of star size. It's really a question of personal tastes, though. Some of your other images already have nice stars; it's just a few of the brightest ones that got away (like in the Dolphin Head image). Some people reduce the star size too far, which is equally bad, in my view. Now that you have a handle on how to manipulate star size during stretching, do what you'd like with it, and don't be too sensitive to other peoples' preferences. The sizes are right when you feel they are right. Couldn't agree more with your comment and now that you mentioned your C11 pictures... you know what I just did! You have some great captures in there. Your version of the cone nebula is great. Gives me a good idea about the resolution in the finer details of the nebula. onwards and upwards!
|
0.00
#...
·
·
1
like
|
|---|
|
I'm a newbie to astrophoto and no jedi, but I'm super picky about which pictures I like, so here's my personal / non-professional feedback: NGC2264: I really like the re-process your did of the NGC2264, feels MUCH more natural and looks way better. Rosette: Beautiful color palette and general look, but the "black ink" looking clouds in the core look way too sharp compared to the rest. It doesn't feel like it fits in. I could be completely wrong, but looks like an overdose of Topaz sharpness or something like that to me. NGC 3372: Love it all overall, looks amazing. Stars do look blurry and no stars spikes. Personal taste, I just like seeing a couple big stars with their spikes. Horsehead: It didn't shock me as much as Taylor haha I find the colors interesting, it gives it a sunset look and I like the blend of colors done. IC 2944: Now this bugs me more for the colors. I don't think the deep red and deep blue compliment each other very well. Compared to your Rosette and Carina, the blend of colors was much richer and spread on these. The contrast and everything else looks good to me tho. Dolphin: Way too much blue. When I edit pictures, the first thing I do after setting the basic colors for each filter is setting a threshold layer and select the blackest pixel I can find not too far off the nebula. In your case, you can see it's way off balance (see below: 39 red, 26 green, 47 blue on one of the darkest pixels I could pick). You probably wanna balance all your colors before choosing specific colors.  I really like this kind of RCC, as you have some solid data and serious editing, plus it helps others (like me!) figure out what they like artistically too. So... thanks for sharing your work and questions Hope this is of any use to you! |
|
10.48
#...
·
|
|---|
I'm a newbie to astrophoto and no jedi, but I'm super picky about which pictures I like, so here's my personal / non-professional feedback: @dreamsandmonsters Thanks so much for taking the time to have a look at my pictures and giving me some really useful feedback. I really appreciate your input! To reply to your comments, Rosette: Yes totally agree and that is indeed an overdose on topaz AI, probably would benefit from a reprocess NGC3372: Thanks for the comment! Don't forget I'm using a refractor, you can't get diffraction spikes with refractor telescope... if that is what you meant...? only reflector or RC telescope due the the spider holding the secondary mirror creating a diffraction spike in the light path. Otherwise I'm using an electronic focuser so the stars shouldn't be blurry and my HFR was around 2.5, it may be a processing thing. Again I have to collect some more data on this one and improve on this a bit Horse-head: This one has a bold colour palette, you either love it or hate it, I'm glad you are not chocked by it! It such a famous target as well people are very used and attached to a certain look. I personally really like the bold flame colour look, hence why I did it this way, it make it pop, chariot of fire style IC 2944: I have collected more data on this and I'm, going to do a complete reprocess soon. I agree the colour are too much in the sense that there is not enough gradient between them. I think the red is probably too intense as well. I was trying to emulate Emuhead version of this target but there has been some great colours processing done on this target recently, actually someone from my Astronomy club won IOTD just a few days ago with this target (https://www.astrobin.com/kr4j7f/C/) and I'm going to sue this as a reference Dolphin: I know what you are saying, the background is far from neutral, and it may be titled too much toward the blue hue and that is probably due to the fact that I was playing with the RGB curves as a last layer and doing a bit of a colour grading to this but also the fact that this target is mostly Oiii and there is a fair amount a faint Oiii signal surround the entire target hence why the background has a fair amount of blue in it as well. My reference was this image from Andy Astro: https://www.astrobin.com/e1iszj/?nc=collection&nce=3940 which has a similar look |
|
0.00
#...
·
·
1
like
|
|---|
Steeve Body: I'm new to astro, so hum... my bad and pardon my ignorance, especially towards gear haha I'm getting star spikes with my refractor currently (without editing). I just read online that it might be due to the cold weather... oh well, the more you know! Steeve Body: What I do in Photoshop is "killing" just enough of the green to start seeing the yellow, then push the yellow towards orange-ish tint, then push the green to be more cyan, push the yellow more, push the green again. Then I go on with the rest of the colors. This way, it's easier to not kill the gradients of colors. You might have lost the variation of colors by pushing them to the extremes too soon? I'd be curious to know your process for colors, I'm still trying to find the best approach. Steeve Body: There was a Youtube tutorial where the picture data of the filters themselves were used as masks on the next layer, inverted and then manipulated with curves to smooth the transition between both layers. I'm not the greatest at explaining so... for example, I use my layers in this order: Ha above, Sii middle, Oiii third. I copy my Oiii data as a mask on my Sii layer and invert. I copy my Sii data as a mask on the Ha layer and invert. This way it might help your other layers "breathe" and let the actual data through. I'll link the video later if you're interested and if I can find it again.  |
|
10.48
#...
·
|
|---|
Steeve Body: Yes please share the video! I haven't really settled on a method for my colors quite yet. The Rosette was done with JP Metsavaino Tone mapping technique: https://www.youtube.com/watch?v=EQlGsM-S8vM and then a lot of my other picture have been done following this workflow and then finalise in photoshop and selective colors to fine tune balance and also love Camera RAW filter for final touches: https://www.youtube.com/watch?v=PArRNIDrae8 I have recently discovered what you are referring to. Literally saw something like this 2 days on the astro imaging channel but I'm yet to use it! Looks very powerful. |
|
0.00
#...
·
·
1
like
|
|---|
|
There you go, I found it. It's from AstroEd Youtube channel: https://www.youtube.com/channel/UC4qgz57tz5eIhjFa9rUcB4Q Channel combine (using the data itself as masks at 19:04) https://www.youtube.com/watch?v=3HYwJNf5M4M Channel separation + denoise(another useful video) https://www.youtube.com/watch?v=tOIzrKA_Ezo&t=1505s I'll take a look at both videos you sent, the more knowledge, the better  I do not have Pixinsight tho, but some ideas can be probably re-done in Photoshop through other methods.Thanks! |
0.90
#...
·
·
1
like
|
|---|
|
False color is the choice of the imager, I really don't have any criticism towards people's color choices here because it is completely up to you. I do think that there is too much noise reduction in some of the images, the dolphin for example. It kind of looks like saran wrap in the background when you use too much NR. I have screwed up everything humanely possible during processing so this is my very anecdotal stance on this issue, haha! Also once you fix the noise I think the details of the nebula will become more dimensional, one thing that can happen while processing is the image can become very flat and structure of the shadows, mid tones, and highlights do not have much distinction. Tonal masking can help you focus on this. Also don't be afraid of allowing some grainy noise to remain in the image, it is a natural characteristic of the data and it helps distinguish the micro contrast regions and details. I think you have some good stuff here, and you clearly have an interest in further developing your talent. |
