Page 3 of 5 FirstFirst 12345 LastLast
Results 21 to 30 of 50
  1. #21
    Flea77's Avatar
    Flea77 is offline SUPER GIANT
    Points: 6,350, Level: 54
    Level completed: 50%, Points required for next Level: 200
    Overall activity: 0%
    Achievements:
    200+ Posts Achievement!400+ Posts AchievementFirst 1000 Experience Points50 Posts Achievement!5 Threads Achievement!
    Join Date
    Aug 2011
    Location
    None
    Posts
    1,277
    Points
    6,350
    Level
    54
    Thanks
    16
    Thanked 775x 415 Posts

    Default



    Hmmm, I see what you mean. I guess I didn't notice that at the dark site, and not even when I posted the above until you pointed it out, it really is just a tad off to the left. Have to watch that in the future, thanks!

    Don't get me wrong, I do like the final image I came up with, I am just completely baffled because I prefer the single processed image to the stacked image. It gets me wondering why I (and everyone else) sit out there for hours capturing 30 lights, 15 darks, 15 bias, and 15 flats if I just need one light and be done with it :-)

    Allan

  2. #22
    carastro's Avatar
    carastro is offline SUPER GIANT
    Points: 18,749, Level: 94
    Level completed: 63%, Points required for next Level: 151
    Overall activity: 7.0%
    Achievements:
    200+ Posts Achievement!First 1000 Experience Points365 Days+ Registered Achievement!400+ Posts Achievement50 Posts Achievement!
    Join Date
    Dec 2010
    Location
    England
    Posts
    951
    Points
    18,749
    Level
    94
    Thanks
    1
    Thanked 636x 281 Posts

    Default

    The focussing mask looks correct, but zooming in will show up any miniscule discrepancies. I have had the same problem on very small objects like M57 which need to be zoomed in on otherwise they are too small.

    There is an even more accurate method you could learn if you wished called Bahtinov grabber. You can use a programme like Nebulosity and then open Bahtinov grabber which grabs a chosen star off the Nebulosity screen and will give you a measurement (+ and -) of how close you are to perfect focus as you are focussing as it continues to take photos of the Bahtinov'd star as you turn the focus wheel.

    I don't think I can add any more to your stacking problem as I don't understand why your single frame should be better than a stacked frame. Perhaps some-one more technical can help.

    Carole

    To view links or images in signatures your post count must be 5 or greater. You currently have 0 signatures.

    Mount: NEQ6 &
    To view links or images in signatures your post count must be 5 or greater. You currently have 0 signatures.

    Imaging
    To view links or images in signatures your post count must be 5 or greater. You currently have 0 signatures.
    : WOZS71 Zenithstar APO, DMK21,
    To view links or images in signatures your post count must be 5 or greater. You currently have 0 signatures.
    Pro ED120, Skywatcher Evostar ED80, SW130PDS
    Camera: Modified Canon EOS 1100D, Arik460EX Mono, Atik314L mono + EFW
    Guiding: Skywatcher finderguider/QHY5 Observatory: SkyPOD + PZT
    Website: https://sites.google.com/site/caroleastroimaging/home

  3. The Following User Says Thank You to carastro For This Useful Post:

    admin (11-04-2011)

  4. #23
    Flea77's Avatar
    Flea77 is offline SUPER GIANT
    Points: 6,350, Level: 54
    Level completed: 50%, Points required for next Level: 200
    Overall activity: 0%
    Achievements:
    200+ Posts Achievement!400+ Posts AchievementFirst 1000 Experience Points50 Posts Achievement!5 Threads Achievement!
    Join Date
    Aug 2011
    Location
    None
    Posts
    1,277
    Points
    6,350
    Level
    54
    Thanks
    16
    Thanked 775x 415 Posts

    Default

    Carole,

    I went and downloaded the Bahtinov Grabber and will try it next time I go out, thanks for that.

    As for the stacking issue, Dave was going to work on it so maybe he will find something. I very much appreciate you testing it out because if nothing else you told me I was not completely crazy, your stacks didn't come up really any different than mine. We may still not have an answer but at least we are not alone, heh.

    Allan

  5. The Following User Says Thank You to Flea77 For This Useful Post:

    admin (11-04-2011)

  6. #24
    carastro's Avatar
    carastro is offline SUPER GIANT
    Points: 18,749, Level: 94
    Level completed: 63%, Points required for next Level: 151
    Overall activity: 7.0%
    Achievements:
    200+ Posts Achievement!First 1000 Experience Points365 Days+ Registered Achievement!400+ Posts Achievement50 Posts Achievement!
    Join Date
    Dec 2010
    Location
    England
    Posts
    951
    Points
    18,749
    Level
    94
    Thanks
    1
    Thanked 636x 281 Posts

    Default

    I went and downloaded the Bahtinov Grabber
    Great, if you have any problems using it, I have the instructions somewhere.

    Carole

    To view links or images in signatures your post count must be 5 or greater. You currently have 0 signatures.

    Mount: NEQ6 &
    To view links or images in signatures your post count must be 5 or greater. You currently have 0 signatures.

    Imaging
    To view links or images in signatures your post count must be 5 or greater. You currently have 0 signatures.
    : WOZS71 Zenithstar APO, DMK21,
    To view links or images in signatures your post count must be 5 or greater. You currently have 0 signatures.
    Pro ED120, Skywatcher Evostar ED80, SW130PDS
    Camera: Modified Canon EOS 1100D, Arik460EX Mono, Atik314L mono + EFW
    Guiding: Skywatcher finderguider/QHY5 Observatory: SkyPOD + PZT
    Website: https://sites.google.com/site/caroleastroimaging/home

  7. #25
    toolmaker's Avatar
    toolmaker is offline Bright Giants
    Points: 10,303, Level: 70
    Level completed: 18%, Points required for next Level: 247
    Overall activity: 0%
    Achievements:
    10 Days registeredFirst 1000 Experience Points365 Days+ Registered Achievement!750 Days+ Registered Achievement!1000 Days+ Registered Achievement!
    Join Date
    Jul 2009
    Location
    Toronto, CAN
    Posts
    536
    Points
    10,303
    Level
    70
    Thanks
    197
    Thanked 355x 212 Posts

    Default

    Sorry for the late response, been busy here.

    From the looks of it, your result from DSS is very similar to what I got using PI.

    First lets take a look at a single unprocessed frame:



    Next we have the result you got from using DSS (stacked but not processed):


    And finally the result I got from using PI (stacked but not processed):


    As you can see the biggest difference between the single frame and the stacked results from DSS and PI is the noise difference. This makes sense since the main reason for stacking many frames together is to increase the SNR of the resulting image. Yes, the stacked images look a bit softer, but I believe that can be from there being less local contrast (small scale changes in brightness, aka noise), and/or from slight differences in the sub frames.

    If we zoom in to 5:1 things become a little more obvious.

    Here's the single frame:


    The result you got from DSS:


    And finally the result from PI:


    At this scale it is very clear that the stacked images have a much better SNR than the single frame, and the stars become rounder and more defined, and the fainter ones separate form the background noise much better. The same goes for the faint nebulosity.

    Give me a little bit more time and I'll try and process the DSS image and my result from PI using the same methods, this will show just how similar the two images really are. Since the stacked images have a much better SNR than the single frame they can be processed much more aggressively, and any softening or star bloating you see in the stacked images can be corrected relatively easily.

    My only concern is how "bright" your result from DSS is. The images I generate from PI are usually much darker in there linear state (before they are stretched) than what you obtained from DSS. I wonder if this is a result of how the subframes were calibrated in DSS, or if there was light pollution at your imaging sight. Is there any way you might be able to share a raw image from your camera, still in the Nikon file format along with the equivalent calibration frames? I'm curious to see the difference between DSS and PI for image calibration.
    Last edited by toolmaker; 11-04-2011 at 01:11 AM.
    Dave
    Celestron 11" EdgeHD
    CGEM DX

    To view links or images in signatures your post count must be 5 or greater. You currently have 0 signatures.
    ST80
    SSAG
    Canon 40D

  8. The Following User Says Thank You to toolmaker For This Useful Post:

    admin (11-04-2011)

  9. #26
    Flea77's Avatar
    Flea77 is offline SUPER GIANT
    Points: 6,350, Level: 54
    Level completed: 50%, Points required for next Level: 200
    Overall activity: 0%
    Achievements:
    200+ Posts Achievement!400+ Posts AchievementFirst 1000 Experience Points50 Posts Achievement!5 Threads Achievement!
    Join Date
    Aug 2011
    Location
    None
    Posts
    1,277
    Points
    6,350
    Level
    54
    Thanks
    16
    Thanked 775x 415 Posts

    Default

    Dave,

    I uploaded a RAW light file into a subdirectory of the "Lights-Tif" dropbox folder you have access to, let me know if you need anything else.

    OK, maybe that is my problem, I am not understanding the terminology here.

    When I think signal to noise ratio I think of sound. 15db means there are 15db between the audio being measured and silence. Following that logic, when someone says stacking increases the SNR I think it increases the difference between the black of space (seemingly black) and the color of the nebula making it stand out from the background more.

    In other words, lets say for the sake of conversation that the background was 250,250,250 and the nebula was 150,150,150 RGB. What I think of as increasing the SNR when the background is basically lightless black is making the background 251,251,251 and the nebula goes to 149,149,149. This is massively oversimplified but you get the drift.

    It sounds like what you are saying is that stacking reduces digital noise, not the SNR. Is that right? If so that would explain everything pretty well from my view but in this day and age of high ISO DSLRs and excellent noise reduction software I would have to wonder why anyone would spend the time shooting hours and hours of images when you could shoot one and run it through a good noise removal program and be done.

    Allan

  10. The Following User Says Thank You to Flea77 For This Useful Post:

    admin (11-04-2011)

  11. #27
    toolmaker's Avatar
    toolmaker is offline Bright Giants
    Points: 10,303, Level: 70
    Level completed: 18%, Points required for next Level: 247
    Overall activity: 0%
    Achievements:
    10 Days registeredFirst 1000 Experience Points365 Days+ Registered Achievement!750 Days+ Registered Achievement!1000 Days+ Registered Achievement!
    Join Date
    Jul 2009
    Location
    Toronto, CAN
    Posts
    536
    Points
    10,303
    Level
    70
    Thanks
    197
    Thanked 355x 212 Posts

    Default

    Thanks for uploading a raw image file Allan.

    I like to think of noise as an uncertainty of signal, and signal is what we're after, photons of some faint object. This may be an incorrect view, I'm not sure.

    I believe that we have a similar definition/understanding of SNR. However, I'd like to point out that stacking does not only reduce the amount of noise in an image, I believe that it also increases the signal in the image. So by reducing the noise and increasing the signal we are able to increase the ratio between these two (SNR) even more than had we just reduced the amount of noise.
    Dave
    Celestron 11" EdgeHD
    CGEM DX

    To view links or images in signatures your post count must be 5 or greater. You currently have 0 signatures.
    ST80
    SSAG
    Canon 40D

  12. The Following User Says Thank You to toolmaker For This Useful Post:

    admin (11-04-2011)

  13. #28
    Flea77's Avatar
    Flea77 is offline SUPER GIANT
    Points: 6,350, Level: 54
    Level completed: 50%, Points required for next Level: 200
    Overall activity: 0%
    Achievements:
    200+ Posts Achievement!400+ Posts AchievementFirst 1000 Experience Points50 Posts Achievement!5 Threads Achievement!
    Join Date
    Aug 2011
    Location
    None
    Posts
    1,277
    Points
    6,350
    Level
    54
    Thanks
    16
    Thanked 775x 415 Posts

    Default

    Quote Originally Posted by toolmaker View Post
    However, I'd like to point out that stacking does not only reduce the amount of noise in an image, I believe that it also increases the signal in the image. So by reducing the noise and increasing the signal we are able to increase the ratio between these two (SNR) even more than had we just reduced the amount of noise.
    THAT is precisely what I am NOT seeing, which is what led me to believe that I must be doing something wrong. When I look at the closeups of the single processed file and the processed stacked file, I see sharper stars (which to me means more detail) and I see more detail in the nebula, specifically in the tops (as the image is oriented). If you disagree and think I am just blind, PLEASE tell me and point out in the image what is better where, I really want to know.

    Now I realize some people stack hundreds of images, but it is hard for me to fathom I can not see an improvement between one and seventy five frames (30 lights + 15 darks + 15 bias + 15 flats) stacked.

    Another interesting point is that when I provide images and information, only two people jumped in to see what they could come up with. Why is that interesting? Well, my understanding in that the vast majority of AP people stack images, so I would have thought that someone would have jumped in by now and said "oh, your problem is ......" and since that has not happened, I am left wondering if anyone really knows why they spend hours and hours gathering images to stack

    Allan

  14. #29
    toolmaker's Avatar
    toolmaker is offline Bright Giants
    Points: 10,303, Level: 70
    Level completed: 18%, Points required for next Level: 247
    Overall activity: 0%
    Achievements:
    10 Days registeredFirst 1000 Experience Points365 Days+ Registered Achievement!750 Days+ Registered Achievement!1000 Days+ Registered Achievement!
    Join Date
    Jul 2009
    Location
    Toronto, CAN
    Posts
    536
    Points
    10,303
    Level
    70
    Thanks
    197
    Thanked 355x 212 Posts

    Default

    Ok, so lets take a look at a 12 to 1 image scale where there is strong signal in the nebula and some stars, and there is weak signal in the background and a few stars.

    Single frame at 12:1


    Unprocessed stack from DSS at 12:1


    At this scale we can see that even though the stars and nebula appear softer they are IMO clearly better defined and their shape/structure are my more noticeable in the weaker signal areas. Not to mention there is a ton less noise.

    There IS better (greater) contrast in the single frame image. That is to say there is a larger difference in brightness between the dark and light areas of the image, such as between the sky background and stars, and between the leading edge of the nebula and the background. However, to me this does not mean better detail or stronger signal. It can appear that way at first, but don't forget these images are both unprocessed and I can guarantee that I can apply much more aggressive processing techniques to the stacked image than to the single frame, making the stacked image look much better than the single frame.

    Why do we stack many frames as opposed to using a single frame?

    1. To increase SNR to a point where we can process an image without introducing unwanted artifacts (very hard to do on a single frame).

    2. To remove unwanted signal such as hot pixels and satellite trails, again very hard to do on a single frame but is almost trivial with pixel rejection during stacking.

    Those are the two big ones, I'm sure there are more reasons but thats all I can think of for now.

    So, to answer your question "whats my problem?", IMO and in the spirit of constructive criticism I think you can work to improve your processing skills (can't we all). Rather than learning "recipes" to process image, try and learn the general concepts that make images look better. Ask yourself what specifically is it that makes a particular image look good, and then work to learn those techniques and apply them when needed.

    I like to break an image down into three parts. The main object, the stars, and the sky background. All three need to be processed differently, there's no point to sharpen the background or increase its colour saturation like you would to a galaxy. Just as you wouldn't want to use as much noise reduction on strong signal areas as you would on weaker areas.

    Practice, practice, practice. One selfish reason that I wanted to try your data was that I've never imaged this target before and I wanted to try processing it (its a tricky object I'm finding, so many bright stars and a relatively faint object). The more images you work on the better feel you'll get for it. Also try and set goals for each image. Before you star try and figure out just what you want to enhance in your image and what it is you'd like to make less obvious.

    This is all just my opinion and I could be completely off base.
    Dave
    Celestron 11" EdgeHD
    CGEM DX

    To view links or images in signatures your post count must be 5 or greater. You currently have 0 signatures.
    ST80
    SSAG
    Canon 40D

  15. The Following 2 Users Say Thank You to toolmaker For This Useful Post:

    admin (11-04-2011),Sc0rch (11-08-2011)

  16. #30
    Flea77's Avatar
    Flea77 is offline SUPER GIANT
    Points: 6,350, Level: 54
    Level completed: 50%, Points required for next Level: 200
    Overall activity: 0%
    Achievements:
    200+ Posts Achievement!400+ Posts AchievementFirst 1000 Experience Points50 Posts Achievement!5 Threads Achievement!
    Join Date
    Aug 2011
    Location
    None
    Posts
    1,277
    Points
    6,350
    Level
    54
    Thanks
    16
    Thanked 775x 415 Posts

    Default

    Quote Originally Posted by toolmaker View Post
    So, to answer your question "whats my problem?", IMO and in the spirit of constructive criticism I think you can work to improve your processing skills (can't we all). Rather than learning "recipes" to process image, try and learn the general concepts that make images look better. Ask yourself what specifically is it that makes a particular image look good, and then work to learn those techniques and apply them when needed.
    I am far too new to this to have a recipe! Having a serious photography background I know that every image may need completely different processing. My problem here is that no matter how I process these images (either in DSS or MaximDL, then in Lightroom, Photoshop and DeNoise) I can not get a better result than the single image. Heck, I can't get as good a finished image IMHO.

    So I am not really looking for a "show me how to process my images" but more a "Look what I can do with your stacks that looks WAY better than your single" so I know it is just me.

    Quote Originally Posted by toolmaker View Post
    I like to break an image down into three parts. The main object, the stars, and the sky background. All three need to be processed differently, there's no point to sharpen the background or increase its colour saturation like you would to a galaxy. Just as you wouldn't want to use as much noise reduction on strong signal areas as you would on weaker areas.
    Do you literally work on each area seperately? If so I would be interested in learning how to break the image into three parts, that is not something I have had any success with. For example, when I try to remove stars what happens is I get them removed, then try compressing the remaining nebula (through levels or curves) and the area around the stars starts to discolor (usually lighten) to the point that when I try to put the stars back in it looks completely off.

    Quote Originally Posted by toolmaker View Post
    Practice, practice, practice. One selfish reason that I wanted to try your data was that I've never imaged this target before and I wanted to try processing it (its a tricky object I'm finding, so many bright stars and a relatively faint object). The more images you work on the better feel you'll get for it. Also try and set goals for each image. Before you star try and figure out just what you want to enhance in your image and what it is you'd like to make less obvious.

    This is all just my opinion and I could be completely off base.
    Oh I absolutely continue to practice. But much like a baseball player, if you do not know what should happen when you hit the ball with the bat or where you should run, I am not sure what the output of my stacked images should be. I would have thought they would be light years ahead of the single images but nothing I do (or anyone else has done either) has managed that. Hence my making this post.

    Allan

 

 
Page 3 of 5 FirstFirst 12345 LastLast

Similar Threads

  1. image stacking
    By darkfield in forum Amateur Astronomy Forum
    Replies: 19
    Last Post: 03-20-2006, 09:48 PM
  2. Image stacking
    By Steve Irwin in forum Amateur Astronomy Forum
    Replies: 4
    Last Post: 06-13-2005, 11:54 PM
  3. Image Stacking
    By Henry in forum Amateur Astronomy Forum
    Replies: 3
    Last Post: 05-24-2004, 03:43 PM
  4. Image stacking
    By Gordon in forum UK Astronomy Forum
    Replies: 1
    Last Post: 12-28-2003, 12:16 PM
  5. Mac image stacking software
    By Juan Hobenaro in forum Amateur Astronomy Forum
    Replies: 0
    Last Post: 09-11-2003, 04:31 AM

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  
Powered by vBulletin® Version 4.2.0
Powered by vBulletin®
All times are GMT. The time now is 05:59 AM.