1.51
#... |
|---|
|
Hello Together! I'm relative new to Astrophotography, started in January 2020. Since the beginning i have improved my Workflow in Pixinsight with Denoising, Deconvolution, Photometric Color Calibration, Curves Transformation, HDRMultiscaleTransform, LRGB Combination. My latest Picture is NGC2204: https://www.astrobin.com/tjku82/B/ From about 50 views, only 10 liked it, that's "only" 20%. But in the case of NGC2204 i thought that this was a really nice Picture, the roundness of the stars, many background Galaxies and good colors. But this seems to be wrong? So how to improve my edit? Thanks Philip |
4.72
#...
·
9
likes
|
|---|
|
Hi Philip ! Nothing is wrong with your NGC2204, maybe just a sky background is a bit too bright. Star shapes and colors are good, details of galaxy are nice. Don't let others to judge your images or your posts anywhere in social media by "likes". This is a great result for someone who have started less than 1 year ago ! If I'll show here my first astrophoto, it will get 100 dislikes.  Best, Nadir |
|
1.51
#... |
|---|
|
Hi Nadir! Thanks for your nice words! I will take a look for the sky background, currently it is about 0.24 of 1.0, what do you think are good values? 0.15? Thanks! Philip |
6.89
#...
·
2
likes
|
|---|
|
Yeah it's a good picture. If you look at your Histogram (shown in the desktop version of the site) it starts very far from the Y axis. So all you have to do is raise the black point a bit . Open Histogram transformation, move the left slider (shadows) a little to the right, bringing the Histogram as close to the Y axis as you like, but without touching the axis. Done. Cheers, Dimitris Ps this is of course what I would do based on the fact that the distance from the Y axis represents a constant value, a glow which has been added to your data and is sheldom real signal). If you prefer the picture the way it is, "20% likes" is not a good reason to change it IMO. If you do not see the glow yourself, you probably need to calibrate your monitor, especially contrast and brightness. |
|
1.51
#...
·
1
like
|
|---|
|
Hey Dimitris, thanks for your advice. I have just uploaded a darker Version: https://www.astrobin.com/tjku82/C/ What do you think? Thanks Philip |
|
4.72
#... |
|---|
Hi Nadir! It's hard to tell by numbers. I agree with Dimitris, just raise a black point a bit till you start to clip some pixels. This will bring the picture to more natural and classic look. Also you can lower stars a bit by MorphologicalTransformation and bring out some galaxy details so that they "pop" a bit. There are many tips and tricks in post-processing, but nothing for you to worry so far or get pissed because nobody "likes" your images. Keep imaging and improve your processing skills.  Nadir |
|
6.89
#... |
|---|
| Much better (for me at least). If you compare the histograms it is obviously closer to the Y axis. |
0.00
#... |
|---|
|
Very good work! If you ask, I would trim a red from a background a little. That would be for my taste only |
11.14
#...
·
2
likes
|
|---|
|
Hello Philip! The image just had a bright red background allowing some minor leftovers from background removal to be visible. All you need to do is to ensure your background is neutral either by a more careful background removal, color calibration, or even by using the BackgroundNeutralization process. Noise-wise the image is fine and I wouldn't bother with star shrinkage or similar things since your priority is the color and brightness of the background sky. Your latest version is towards the right direction, but if you examine the histogram, red still dominates the image |
|
1.51
#... |
|---|
|
Thanks for that hint with the Red Channel, i have adjusted it: https://www.astrobin.com/tjku82/D/ CS Philip |
3.01
#...
·
1
like
|
|---|
|
Hi Philip Rev C is much better, and maybe you should correct a bit the colour balance. You have a bit too much red in the image , at least on my monitor. I downloaded your image and I adjusted a bit the histogram and colour. Philip BTW a "like"-percentage between 20 - 30% is pretty common here on Astrobin😊 cs johny |
 #... |
|---|
|
Hi there! I read just now the post and various answer given. I agree with most of them, but I think that the problem with red channel is due to SCNR with an amount too high. Maybe I’m wrong but the image appear to me more on Magenta then pure Red and this is basically a lack of green. I also suggest you to use a star mask with SCNR because it stole the “green part of yellow” from your stars which are now all orange-red even if a high percentage of stars in universe are actually yellow. You can take a correction now by using a star mask and curves transformation, adjusting oranges to yellow in the Hue page. P.s. it is a very very good photo, I appreciate the fact you use your CMOS at gain 0, but I think 180” subframes are quite short. Remember: longer subframes = more signal. Exploit your bortle 4 sky |
|
1.51
#...
·
1
like
|
|---|
|
Hey Alessio! Thanks for your comment! I will take care about the SCNR in future! Thanks! Yes i'm using Gain-0, usualy i do 300s Subs for Narrowband and 180s for Broadband. I will give longer Broadband Subs a try in future Thanks! Philip |
2.71
#...
·
1
like
|
|---|
| Nothing - I actually like the less contrasty versions, B is excellent, more 'filmic'. All are good! |
0.90
#... |
|---|
|
Hi Philip, Great capture of NGC2403 there. My favourite version is "D" by far. The colors of your galaxy really shine in that version. My only remark would be that some of the stars are a little over-saturated. Personally, I don't mind that. I think it is a style and adds some romanticism. May I ask if you are editing your images on a calibrated monitor that is suitable for the job? I hate to be that guy on the forum that points it out. But when I read the feedback above, and see the new versions you upload, I get the feeling that the difference between what you see and what I see might be significant. Perhaps check how your image looks on a couple of different monitors and a smartphone or two. You are obviously talented so I believe it won't take long before these adjustments will feel like second nature. I'm speaking from daytime experience here, as an astrophotographer I'm pretty fresh myself. Looking forward to your next one. |
|
1.51
#... |
|---|
|
Thank you, no i don't have a calibrated Monitor. But i figured out, that my Laptop Monitor displays the colors / contrast very different then my regular Monitor. So it is i bit tricky to find out, how "others" will see my pictures. I guess that the most poeple don't have a calibrated monitor? CS Philip |
18.84
#... |
|---|
I guess that the most poeple don't have a calibrated monitor? Hi Philip, I think there is no way around to use a calibrated display. I use a "Spyder 5 Elite" to calibrate it every 8 weeks together with a monitor which is made for image processing. I have tried before without and it was an epic fail. Actually color calibration is essential. CS Rüdiger |
|
0.90
#...
·
3
likes
|
|---|
|
Hi Philip, I used to teach colorgrading classes (video), and I recognised some of the conversations. The good news is that your images probably looked great on your monitor the first time around. And all the suggestions people made will feel more natural when you look at them on a calibrated monitor, especially if you are currently using a laptop screen. Perhaps it is not a bad screen, but with laptops, it is tough to keep a consistent viewing angle. Just try and move your head up and down a little, your image probably changes considerably. You are right. Most people don’t calibrate their monitors. Creators perhaps but probably not the people who are browsing the web looking at your image. Allow me to quickly explain why calibration is still essential: Imagine a big target or a dartboard. In the centre, there is a bulls-eye this is the precise colorspace in which your image is being displayed online (sRGB). The closer your monitor comes to displaying sRGB the closer you are to the bulls-eye. We measure the difference with a DeltaE value. When DeltaE is smaller than 2, the difference is imperceivable. Professional colorists aim to have a DeltaE <2 for both Gamma (contrast) and Gamut(colors). Now you might have a DeltaE of 6, which is not bad in itself. But your DeltaE might be one way, towards green. This will lead you to remove more green than necessary, and your image turns out magenta. What’s more, someone else who sees your photo might have a DeltaE of 8. Again, not terrible. But if that DeltaE leans the other way, towards red, the difference between what you see and what that other person sees could be DeltaE 14. That is a big difference. Lastly, I’ve noticed that images that are edited on an accurate monitor, look better on bad monitors. I think it has to do with the fact that each monitor is expected to display a good sRGB image. These days, you really don’t have to obsess over calibration. Once calibrated, they really don’t drift as much as they used to (dependent on the technology behind the panel). But it definitely does pay to choose a monitor that is known for good colour reproduction. Some of them even come pretty well-calibrated from the factory. I see that some entry-level EIZO monitor currently goes for E315- here in the Netherlands. The more fancy ones even have a calibrator build in. I myself use an HP Dreamcolor that I calibrate about twice a year. (My professional monitors I calibrate more often but that’s work). I know it’s a pain but in the end it’s just another piece of kitt that’s part of the hobby. Any questions, let me know. you are on the right track. Regards, Remco |
1.43
#...
·
3
likes
|
|---|
|
I am by no means an expert, so take this with a grain of salt, however in the darker versions — especially the ‘final’ I see that the edges of the photo are much darker than the center (this is very apparent, on my iPad, at the right edge). To me this looks like the image might not have been cropped correctly in PI before color calibration/background extraction — or at least that’s the kind of weirdness that I see in my own photos when I’m trying to preserve as many pixels as possible and miss the crop by a few pixels. I’ll echo what others have said, this is a great photo and likes is a bad measure of your skill! |
|
1.51
#... |
|---|
|
Thanks for your Tipps and your suggestions. @Clint i have cropped the image. The Asi1600 has a resolution of 4656x3520, this image has 4520x3376. @Remco Thanks for that info, i've never thought about it any further. I'm editing my Pictures 99% on my laptop. Do you now any goot method to calibrate a display without any tools  ?CS Philip |
|
18.84
#... |
|---|
|
Hello Philip, there are some tools, but they will only help to adjust contrast and brightness. For color you will need an objective and neutral sensors, which the human eye is not. There is no alternative to hardware. As stated above, a laptop monitor is fairly the worst choice to use for image processing. 99% are made for office tasks and have very poor DTP capabilities. Also if you move the laptop a little bit or tilt the lid the light angle on the screen changes. That is the reason why DTP screens are fixed and shaded from ambient light. In professional areas the ambient light is also calibrated, since changes e.g. from daylight to artificial light void the calibration. And this is not only a minor effect, but huge one. E.g. when processing images during day, they look completely different in the evening with artificial light. This is related to the behavior of the human eye. It gets used to certain color temperatures. The human eye starts to compensate shifts in the spectrum (color temperature) over the time. It is per design not usable for such calibrations, since it already tries to calibrate itself. Long story short: Without calibration hardware color adjustment is pure gambling. Any color calibration done without hardware will always be perceived completely different by different people. CS Rüdiger |
3.82
#...
·
1
like
|
|---|
| you've got a great image there and some great feedback/suggestions! i'll just add this. sometimes you can have a really awesome image and it doesn't get the attention it deserves, or what you might like to get! ;). that doesn't mean its not an awesome image. lots of variables as to how that image is seen on the web. From time of day you post to how the image looks as a thumbnail compared to all the other tumbnails next to it! it can be frustrating for sure. |
11.07
#...
·
1
like
|
|---|
Hi Nadir! .15 might be right but let your eye judge. You might want to increase saturation a bit after making that change but I’d wait and see about color till you lower the background level first. |
2.11
#...
·
1
like
|
|---|
|
Hey pmneo ! I like your pic, I am even newer to the “hobby” as I started in November last year so probably I’m the last that should give advice. I played a lot at gain 0 (well... 20...) and 300s subs since I moved from a short period with a NikonD5600, mostly I think for some kind of inner bias that recites “astropics should be loooooong exposures”. then I asked myself why they (CMOS cameras manufacturers) should put such a wide range of gain and offset to be used if then the aim should be to keep them at 0... I have then read some documentation online and discovered that they can actually improve your pictures if well used. And just tonight I shot my first pic at “unity gain” settings ... and I’m very eager to process it (shooting darks now...) as subs looks better already uncalibrated and unstretched. By my absolute beginner eye what I noticed is that you hardly reach a “black” sky, probably because trying to move the histogram you have noticed that you are also losing part of the galaxy... been there several times... frustrating... Maybe try to have some gain and offset, at 2/3 mins subs in LRGB. I think it could help in postprocessing. enjoy |
|
1.51
#... |
|---|
|
Thank you all for your useful Tips! I have produced some more Galaxy Pictures with the new knowledge: M106: https://www.astrobin.com/ad5yi2/?nc=user M51: https://www.astrobin.com/8c2v0g/?nc=user M101: https://www.astrobin.com/8fa3xi/D/?nc=user And M81 / M82: https://www.astrobin.com/2a8d1d/I/?nc=user with one crops of M82: https://www.astrobin.com/s34g95/?nc=user CS Philip |
