Back to forums

HDR M42 (Bortle 8-9)

Optolong L-Pro ZWO ASI290MC Beginner Request for feedback
6 replies•634 views
@RUBYC111@ avatar
@RUBYC111@ avatar
Hello 
I am starting with astrophotography last year.
Please could you give some advance in order to improve my performance?? I will appreciate you frankly opinion.


https://astrob.in/9a8j0x/0/
Main picture parameter: 
HDR 10sg/60sg/180sg
Telescopio: Skywatcher 150/750
Montura: Skywatcher HEQ5
Camera: ZWO Asi 294 MC PRO
Guide: Skywatcher 50ED  Guide camera ZwoAsi 290MC
Focuser: ZWO EAF
Filter: Optolong L-Pro
Sky : Bortle 8-9
Adquisición NINA
Procesado: PIX INSIGHT
D. Jung avatar
D. Jung avatar
3 main things immediately come up here:

- Your image is very soft, you could make more use of  unsharp and advsharpen (and local histogram normalization) or be more subtle on the denoise in the first step

- your contrast is quite low, you could use HDR multiscaletransform and local histogram normalization to fix that (the core of your m42 is completely flat)

- with the top 2 things addressed it should be easier to pop up the colors
Helpful Concise
robonrome avatar
robonrome avatar
hi ruby, it looks like you have some good data to start, but hard to give good guidance without more information about what you have done.

You've titled this HDR, did you take different expsoure lengths to try and preserve the core?  What exposure lengths and how did you blend these for the HDR?

You've managed to preserve the trapezium stars, but as the poster above notes its very flat with no central detail… hard to say why and how to improve without more info as above.

The above aside, I think the main issue here is too much noise reduction…it's way over-smoothed and lacking detail as a result.

My advice. You're better off tolerating some noise and retaining detail. Of course getting more total exposure time will help you better manage noise and with bortle 8-9 you are in some very challenging conditions…. needs lots and lots (hundeeds) of shorter exposures in general to  build up the integration and with an object like m42 probably needs a range of exposures to bring out fainter detail while preserving core…e.g. as little as 15s expsoures up to a couple of minutes.

Good luck.

rob
Helpful
@RUBYC111@ avatar
@RUBYC111@ avatar
D. Jung:
3 main things immediately come up here:

- Your image is very soft, you could make more use of  unsharp and advsharpen (and local histogram normalization) or be more subtle on the denoise in the first step

- your contrast is quite low, you could use HDR multiscaletransform and local histogram normalization to fix that (the core of your m42 is completely flat)

- with the top 2 things addressed it should be easier to pop up the colors

Hi Jung  (sorry my English could be no so good).
Thank you for you comet and advise. My understanding is that main concern is in M42 core,  isn't it.?
What do you means with  "use of  unsharp and advsharpen"?
Let me explain PI work flow in thsi picture: Chanel extraction/linear fit (refernce imagen R chanel)/Chanel combination/DBE (subtraction) /DBE (division) /Background neutralization/SNCR then I used EZ Denoise (standard parameters) before stretch, StartEXtermination after that in lineal fase I used EZ HDR with parameter Blend 0.15 & HDR layer 3, other combination game me a result very estrange ( some extrange "mointains" instead smooth transitions) ... After that LocalHistoEqua with 2 diferents Lightness mask generate with Range selection) some curve adjust + finally add starts again.

Could be I have to check other noise redution method??
Is it correct/aceptable work flow??

Thank you again for you collaboration & support, it is very appreciate from mi side.

Best regards & clear skies!
D. Jung avatar
D. Jung avatar
EZ denoise sometimes overshoots a bit, resulting in very soft image. Not sure that's your problem here, but regardless of that, you usually need to apply some sharpening filters afterwards like "unsharpen" or "advsharp", the latter of which is part ofa plugin you would 1need to install in PI first.

When you do the HDR combination, you get an image with a very high dynamic range (32 or 64bit) but what you see on the screen is only 8bit, maybe 10 bit if you got an HDR screen. So you need to do some clever histogram operations to not kill the contrast after applying your global histogram stretch, which is some sort of gamma function. Most common fix is too apply local histogram normalization, but pi also offeres a tool called HDR multiscale linear transform which helps you with that. Just experiment a little bit and see what it does 


Here is one of my shots of m42 under bortle 8, only 40minutes and no HDR.

https://www.astrobin.com/3ey9pt/

It's nothing great, but you can see there is a lot of detail in the core that you're missing, which you should be able to get even under light polluted skies.
Helpful Engaging Supportive
@RUBYC111@ avatar
@RUBYC111@ avatar
D. Jung:
EZ denoise sometimes overshoots a bit, resulting in very soft image. Not sure that's your problem here, but regardless of that, you usually need to apply some sharpening filters afterwards like "unsharpen" or "advsharp", the latter of which is part ofa plugin you would 1need to install in PI first.

When you do the HDR combination, you get an image with a very high dynamic range (32 or 64bit) but what you see on the screen is only 8bit, maybe 10 bit if you got an HDR screen. So you need to do some clever histogram operations to not kill the contrast after applying your global histogram stretch, which is some sort of gamma function. Most common fix is too apply local histogram normalization, but pi also offeres a tool called HDR multiscale linear transform which helps you with that. Just experiment a little bit and see what it does 


Here is one of my shots of m42 under bortle 8, only 40minutes and no HDR.

https://www.astrobin.com/3ey9pt/

It's nothing great, but you can see there is a lot of detail in the core that you're missing, which you should be able to get even under light polluted skies.


I have tried again based in your comets, I have tried to reduce brightness in core and increase the core colours used color mask, could be this  version is better than the previous one?
@RUBYC111@ avatar
@RUBYC111@ avatar
hi ruby, it looks like you have some good data to start, but hard to give good guidance without more information about what you have done.

You've titled this HDR, did you take different expsoure lengths to try and preserve the core?  What exposure lengths and how did you blend these for the HDR?

You've managed to preserve the trapezium stars, but as the poster above notes its very flat with no central detail... hard to say why and how to improve without more info as above.

The above aside, I think the main issue here is too much noise reduction...it's way over-smoothed and lacking detail as a result.

My advice. You're better off tolerating some noise and retaining detail. Of course getting more total exposure time will help you better manage noise and with bortle 8-9 you are in some very challenging conditions.... needs lots and lots (hundeeds) of shorter exposures in general to  build up the integration and with an object like m42 probably needs a range of exposures to bring out fainter detail while preserving core...e.g. as little as 15s expsoures up to a couple of minutes.

Good luck.

rob

Hi Rob.
Thank you for you comment, let me explain better about HDR composition lights/Darks/Flats& Darkflats..Optolon L-Pro: 116x10" (19' 20") (gain: 121.00) -20C bin 1x1
Optolon L-Pro: 23x180" (1h 9') (gain: 121.00) -20C bin 1x1
Optolon L-Pro: 57x60" (57') (gain: 121.00) -20C bin 1x1Integración time:2h 25' 20"Darks:60Flats:60Dark-flats:60
According your comments and other "astrobiner" I have done a new process of M42 HDR trying to keep core colors & reduce core brigthness....
Could be thsi new version has been improved related with previous one??

Thank you again for you comment and support, I really appreciate you collaboration and advises.
Helpful Respectful Supportive